drgreen/android_frameworks_base
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
android_frameworks_base/services/input/InputReader.cpp

6278 lines
241 KiB
C++
Raw Normal View History

Mouse pointer integration. Added support for loading the pointer icon from a resource. Moved the system server related bits of the input manager out of libui and into libinput since they do not need to be linked into applications. Change-Id: Iec11e0725b3add2b905c51f8ea2c3b4b0d1a2d67
2011-01-02 16:37:43 -08:00
/*
* Copyright (C) 2010 The Android Open Source Project
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
#define LOG_TAG "InputReader"
//#define LOG_NDEBUG 0
// Log debug messages for each raw event received from the EventHub.
#define DEBUG_RAW_EVENTS 0
// Log debug messages about touch screen filtering hacks.
Native input event dispatching. Target identification is now fully native. Fixed a couple of minor issues related to input injection. Native input enabled by default, can be disabled by setting WindowManagerPolicy.ENABLE_NATIVE_INPUT_DISPATCH to false. Change-Id: I7edf66ed3e987cc9306ad4743ac57a116af452ff
2010-06-22 01:27:15 -07:00
#define DEBUG_HACKS 0
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
// Log debug messages about virtual key processing.
Native input event dispatching. Target identification is now fully native. Fixed a couple of minor issues related to input injection. Native input enabled by default, can be disabled by setting WindowManagerPolicy.ENABLE_NATIVE_INPUT_DISPATCH to false. Change-Id: I7edf66ed3e987cc9306ad4743ac57a116af452ff
2010-06-22 01:27:15 -07:00
#define DEBUG_VIRTUAL_KEYS 0
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
// Log debug messages about pointers.
Disable pointer debug logging. Change-Id: I00628fa369cb1b179a1907f9cb19de8ca4c755ab
2011-05-24 14:40:35 -07:00
#define DEBUG_POINTERS 0
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Even more native input dispatch work in progress. Added more tests. Fixed a regression in Vector. Fixed bugs in pointer tracking. Fixed a starvation issue in PollLoop when setting or removing callbacks. Fixed a couple of policy nits. Modified the internal representation of MotionEvent to be more efficient and more consistent. Added code to skip/cancel virtual key processing when there are multiple pointers down. This helps to better disambiguate virtual key presses from stray touches (such as cheek presses). Change-Id: I2a7d2cce0195afb9125b23378baa94fd2fc6671c
2010-06-16 01:53:36 -07:00
// Log debug messages about pointer assignment calculations.
#define DEBUG_POINTER_ASSIGNMENT 0
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Log debug messages about gesture detection.
#define DEBUG_GESTURES 0
Add support for input devices that have vibrators. Added a getVibrator() method to InputDevice which returns a Vibrator associated with that input device. Its uses the same API as the system vibrator which makes it easy for applications to be modified to use one or the other. Bug: 6334179 Change-Id: Ifc7f13dbcb778670f3f1c07ccc562334e6109d2e
2012-04-13 04:09:27 -07:00
// Log debug messages about the vibrator.
#define DEBUG_VIBRATOR 0
Mouse pointer integration. Added support for loading the pointer icon from a resource. Moved the system server related bits of the input manager out of libui and into libinput since they do not need to be linked into applications. Change-Id: Iec11e0725b3add2b905c51f8ea2c3b4b0d1a2d67
2011-01-02 16:37:43 -08:00
#include "InputReader.h"
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
#include <cutils/log.h>
frameworks/base refactoring. First step. Move libui includes to their new home: androidfw. Change-Id: Ic042b52fdba72f30edc3cc6339bf30b4c1b99662
2012-02-17 15:34:57 -08:00
#include <androidfw/Keyboard.h>
#include <androidfw/VirtualKeyMap.h>
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
#include <stddef.h>
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
#include <stdlib.h>
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
#include <unistd.h>
#include <errno.h>
#include <limits.h>
Add support for new input sources. Added several new coordinate values to MotionEvents to capture touch major/minor area, tool major/minor area and orientation. Renamed NDK input constants per convention. Added InputDevice class in Java which will eventually provide useful information about available input devices. Added APIs for manufacturing new MotionEvent objects with multiple pointers and all necessary coordinate data. Fixed a bug in the input dispatcher where it could get stuck with a pointer down forever. Fixed a bug in the WindowManager where the input window list could end up containing stale removed windows. Fixed a bug in the WindowManager where the input channel was being removed only after the final animation transition had taken place which caused spurious WINDOW DIED log messages to be printed. Change-Id: Ie55084da319b20aad29b28a0499b8dd98bb5da68
2010-07-14 18:48:53 -07:00
#include <math.h>
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
#define INDENT " "
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
#define INDENT2 " "
#define INDENT3 " "
#define INDENT4 " "
Add some logging code to track down touch issues. Change-Id: I9bd7af6b93f82a83c6863986d21e7a222d97e4ae
2011-06-28 20:34:40 -07:00
#define INDENT5 " "
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
namespace android {
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// --- Constants ---
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
// Maximum number of slots supported when using the slot-based Multitouch Protocol B.
static const size_t MAX_SLOTS = 32;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
// --- Static Functions ---
template<typename T>
inline static T abs(const T& value) {
return value < 0 ? - value : value;
}
template<typename T>
inline static T min(const T& a, const T& b) {
return a < b ? a : b;
}
Even more native input dispatch work in progress. Added more tests. Fixed a regression in Vector. Fixed bugs in pointer tracking. Fixed a starvation issue in PollLoop when setting or removing callbacks. Fixed a couple of policy nits. Modified the internal representation of MotionEvent to be more efficient and more consistent. Added code to skip/cancel virtual key processing when there are multiple pointers down. This helps to better disambiguate virtual key presses from stray touches (such as cheek presses). Change-Id: I2a7d2cce0195afb9125b23378baa94fd2fc6671c
2010-06-16 01:53:36 -07:00
template<typename T>
inline static void swap(T& a, T& b) {
T temp = a;
a = b;
b = temp;
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
inline static float avg(float x, float y) {
return (x + y) / 2;
}
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
inline static float distance(float x1, float y1, float x2, float y2) {
return hypotf(x1 - x2, y1 - y2);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Add support for detecting finger orientation. Change-Id: I901fc61f7705fcd7950190dc8782bad616032ee2
2011-01-14 19:09:23 -08:00
inline static int32_t signExtendNybble(int32_t value) {
return value >= 8 ? value - 16 : value;
}
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
static inline const char* toString(bool value) {
return value ? "true" : "false";
}
Fix off by one errors in touch motion ranges. (DO NOT MERGE) Report inclusive minimum and maximum ranges for all axes including X and Y. Set mouse pointer bounds to 0..width-1, 0..height-1. Rotate touch and mouse positions more carefully, paying attention to the maximum bounds when calculating the complement of an axis. Simplified the InputReader somewhat and removed support for a couple of poorly defined input device configuration parameters. We now assume that the touch device provides useful absolute axis ranges for the X and Y axes since the alternative does not actually make sense. Bug: 3413541 Change-Id: I121d28a125c4f9618cb283dc460d33ff1a907023
2011-03-03 02:09:54 -08:00
static int32_t rotateValueUsingRotationMap(int32_t value, int32_t orientation,
const int32_t map[][4], size_t mapSize) {
if (orientation != DISPLAY_ORIENTATION_0) {
for (size_t i = 0; i < mapSize; i++) {
if (value == map[i][0]) {
return map[i][orientation];
}
}
}
return value;
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
static const int32_t keyCodeRotationMap[][4] = {
// key codes enumerated counter-clockwise with the original (unrotated) key first
// no rotation, 90 degree rotation, 180 degree rotation, 270 degree rotation
Add initial gamepad support. Change-Id: I0439648f6eb5405f200e4223c915eb3a418b32b9
2010-06-30 16:10:35 -07:00
{ AKEYCODE_DPAD_DOWN, AKEYCODE_DPAD_RIGHT, AKEYCODE_DPAD_UP, AKEYCODE_DPAD_LEFT },
{ AKEYCODE_DPAD_RIGHT, AKEYCODE_DPAD_UP, AKEYCODE_DPAD_LEFT, AKEYCODE_DPAD_DOWN },
{ AKEYCODE_DPAD_UP, AKEYCODE_DPAD_LEFT, AKEYCODE_DPAD_DOWN, AKEYCODE_DPAD_RIGHT },
{ AKEYCODE_DPAD_LEFT, AKEYCODE_DPAD_DOWN, AKEYCODE_DPAD_RIGHT, AKEYCODE_DPAD_UP },
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
};
Fix off by one errors in touch motion ranges. (DO NOT MERGE) Report inclusive minimum and maximum ranges for all axes including X and Y. Set mouse pointer bounds to 0..width-1, 0..height-1. Rotate touch and mouse positions more carefully, paying attention to the maximum bounds when calculating the complement of an axis. Simplified the InputReader somewhat and removed support for a couple of poorly defined input device configuration parameters. We now assume that the touch device provides useful absolute axis ranges for the X and Y axes since the alternative does not actually make sense. Bug: 3413541 Change-Id: I121d28a125c4f9618cb283dc460d33ff1a907023
2011-03-03 02:09:54 -08:00
static const size_t keyCodeRotationMapSize =
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
sizeof(keyCodeRotationMap) / sizeof(keyCodeRotationMap[0]);
Remove more edge flag cruft. Change-Id: I4f9f4f4cc7c771a918aee82a941a3e86ecfe48fe
2011-07-15 19:08:26 -07:00
static int32_t rotateKeyCode(int32_t keyCode, int32_t orientation) {
Fix off by one errors in touch motion ranges. (DO NOT MERGE) Report inclusive minimum and maximum ranges for all axes including X and Y. Set mouse pointer bounds to 0..width-1, 0..height-1. Rotate touch and mouse positions more carefully, paying attention to the maximum bounds when calculating the complement of an axis. Simplified the InputReader somewhat and removed support for a couple of poorly defined input device configuration parameters. We now assume that the touch device provides useful absolute axis ranges for the X and Y axes since the alternative does not actually make sense. Bug: 3413541 Change-Id: I121d28a125c4f9618cb283dc460d33ff1a907023
2011-03-03 02:09:54 -08:00
return rotateValueUsingRotationMap(keyCode, orientation,
keyCodeRotationMap, keyCodeRotationMapSize);
}
Add support for orientationAware touch pads. Bug: 5038388 Change-Id: I9c571b1542299334d2d64c5b33e456ec20ebc0ae
2011-07-15 20:44:17 -07:00
static void rotateDelta(int32_t orientation, float* deltaX, float* deltaY) {
float temp;
switch (orientation) {
case DISPLAY_ORIENTATION_90:
temp = *deltaX;
*deltaX = *deltaY;
*deltaY = -temp;
break;
case DISPLAY_ORIENTATION_180:
*deltaX = -*deltaX;
*deltaY = -*deltaY;
break;
case DISPLAY_ORIENTATION_270:
temp = *deltaX;
*deltaX = -*deltaY;
*deltaY = temp;
break;
}
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
static inline bool sourcesMatchMask(uint32_t sources, uint32_t sourceMask) {
return (sources & sourceMask & ~ AINPUT_SOURCE_CLASS_MASK) != 0;
}
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
// Returns true if the pointer should be reported as being down given the specified
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
// button states. This determines whether the event is reported as a touch event.
static bool isPointerDown(int32_t buttonState) {
return buttonState &
(AMOTION_EVENT_BUTTON_PRIMARY | AMOTION_EVENT_BUTTON_SECONDARY
Hover and stylus API cleanup. Added missing setOnHoverListener method. Removed BUTTON_STATE_ERASER. Change-Id: I5d820d0a8e47fe5fd2d88e9fd9fc3fc6ce0dea0d
2011-06-27 18:36:00 -07:00
| AMOTION_EVENT_BUTTON_TERTIARY);
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
}
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
static float calculateCommonVector(float a, float b) {
if (a > 0 && b > 0) {
return a < b ? a : b;
} else if (a < 0 && b < 0) {
return a > b ? a : b;
} else {
return 0;
}
}
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
static void synthesizeButtonKey(InputReaderContext* context, int32_t action,
nsecs_t when, int32_t deviceId, uint32_t source,
uint32_t policyFlags, int32_t lastButtonState, int32_t currentButtonState,
int32_t buttonState, int32_t keyCode) {
if (
(action == AKEY_EVENT_ACTION_DOWN
&& !(lastButtonState & buttonState)
&& (currentButtonState & buttonState))
|| (action == AKEY_EVENT_ACTION_UP
&& (lastButtonState & buttonState)
&& !(currentButtonState & buttonState))) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
NotifyKeyArgs args(when, deviceId, source, policyFlags,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
action, 0, keyCode, 0, context->getGlobalMetaState(), when);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
context->getListener()->notifyKey(&args);
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
}
}
static void synthesizeButtonKeys(InputReaderContext* context, int32_t action,
nsecs_t when, int32_t deviceId, uint32_t source,
uint32_t policyFlags, int32_t lastButtonState, int32_t currentButtonState) {
synthesizeButtonKey(context, action, when, deviceId, source, policyFlags,
lastButtonState, currentButtonState,
AMOTION_EVENT_BUTTON_BACK, AKEYCODE_BACK);
synthesizeButtonKey(context, action, when, deviceId, source, policyFlags,
lastButtonState, currentButtonState,
AMOTION_EVENT_BUTTON_FORWARD, AKEYCODE_FORWARD);
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// --- InputReaderConfiguration ---
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
bool InputReaderConfiguration::getDisplayInfo(bool external, DisplayViewport* outViewport) const {
const DisplayViewport& viewport = external ? mExternalDisplay : mInternalDisplay;
if (viewport.displayId >= 0) {
*outViewport = viewport;
return true;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
}
return false;
}
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
void InputReaderConfiguration::setDisplayInfo(bool external, const DisplayViewport& viewport) {
DisplayViewport& v = external ? mExternalDisplay : mInternalDisplay;
v = viewport;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
// --- InputReader ---
InputReader::InputReader(const sp<EventHubInterface>& eventHub,
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
const sp<InputReaderPolicyInterface>& policy,
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
const sp<InputListenerInterface>& listener) :
mContext(this), mEventHub(eventHub), mPolicy(policy),
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
mGlobalMetaState(0), mGeneration(1),
mDisableVirtualKeysTimeout(LLONG_MIN), mNextTimeout(LLONG_MAX),
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
mConfigurationChangesToRefresh(0) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mQueuedListener = new QueuedInputListener(listener);
{ // acquire lock
AutoMutex _l(mLock);
refreshConfigurationLocked(0);
updateGlobalMetaStateLocked();
} // release lock
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
InputReader::~InputReader() {
for (size_t i = 0; i < mDevices.size(); i++) {
delete mDevices.valueAt(i);
}
}
void InputReader::loopOnce() {
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
int32_t oldGeneration;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
int32_t timeoutMillis;
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
bool inputDevicesChanged = false;
Vector<InputDeviceInfo> inputDevices;
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
{ // acquire lock
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
AutoMutex _l(mLock);
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
oldGeneration = mGeneration;
timeoutMillis = -1;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t changes = mConfigurationChangesToRefresh;
if (changes) {
mConfigurationChangesToRefresh = 0;
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
timeoutMillis = 0;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
refreshConfigurationLocked(changes);
Add support for input devices that have vibrators. Added a getVibrator() method to InputDevice which returns a Vibrator associated with that input device. Its uses the same API as the system vibrator which makes it easy for applications to be modified to use one or the other. Bug: 6334179 Change-Id: Ifc7f13dbcb778670f3f1c07ccc562334e6109d2e
2012-04-13 04:09:27 -07:00
} else if (mNextTimeout != LLONG_MAX) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
nsecs_t now = systemTime(SYSTEM_TIME_MONOTONIC);
timeoutMillis = toMillisecondTimeoutDelay(now, mNextTimeout);
}
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
} // release lock
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
size_t count = mEventHub->getEvents(timeoutMillis, mEventBuffer, EVENT_BUFFER_SIZE);
Add a preference panel for mouse speed. Bug: 4124987 Change-Id: I3ce175d268a1d043cf5878481261b1049a15a149
2011-06-02 01:26:32 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
{ // acquire lock
AutoMutex _l(mLock);
Improve watchdog monitor for InputReader and InputDispatcher. Try harder to test for liveness. There are situations where the lock might not be held but the input system is stuck in a callback into the window manager policy that has hung. Bug: 5094994 Change-Id: Iff88655512a5dc8bbb4615be65f4115e975c020b
2012-01-27 17:32:06 -08:00
mReaderIsAliveCondition.broadcast();
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (count) {
processEventsLocked(mEventBuffer, count);
}
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
if (mNextTimeout != LLONG_MAX) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
nsecs_t now = systemTime(SYSTEM_TIME_MONOTONIC);
Improve watchdog monitor for InputReader and InputDispatcher. Try harder to test for liveness. There are situations where the lock might not be held but the input system is stuck in a callback into the window manager policy that has hung. Bug: 5094994 Change-Id: Iff88655512a5dc8bbb4615be65f4115e975c020b
2012-01-27 17:32:06 -08:00
if (now >= mNextTimeout) {
Refactor how timeouts are calculated. Added a timeout mechanism to EventHub and InputReader so that InputMappers can request timeouts to perform delayed processing of input when needed. Change-Id: Iec2045baaf4e67690b15eef3c09a58d5cac76897
2011-03-17 01:34:19 -07:00
#if DEBUG_RAW_EVENTS
Improve watchdog monitor for InputReader and InputDispatcher. Try harder to test for liveness. There are situations where the lock might not be held but the input system is stuck in a callback into the window manager policy that has hung. Bug: 5094994 Change-Id: Iff88655512a5dc8bbb4615be65f4115e975c020b
2012-01-27 17:32:06 -08:00
ALOGD("Timeout expired, latency=%0.3fms", (now - mNextTimeout) * 0.000001f);
Refactor how timeouts are calculated. Added a timeout mechanism to EventHub and InputReader so that InputMappers can request timeouts to perform delayed processing of input when needed. Change-Id: Iec2045baaf4e67690b15eef3c09a58d5cac76897
2011-03-17 01:34:19 -07:00
#endif
Improve watchdog monitor for InputReader and InputDispatcher. Try harder to test for liveness. There are situations where the lock might not be held but the input system is stuck in a callback into the window manager policy that has hung. Bug: 5094994 Change-Id: Iff88655512a5dc8bbb4615be65f4115e975c020b
2012-01-27 17:32:06 -08:00
mNextTimeout = LLONG_MAX;
timeoutExpiredLocked(now);
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
if (oldGeneration != mGeneration) {
inputDevicesChanged = true;
getInputDevicesLocked(inputDevices);
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
} // release lock
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
// Send out a message that the describes the changed input devices.
if (inputDevicesChanged) {
mPolicy->notifyInputDevicesChanged(inputDevices);
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// Flush queued events out to the listener.
// This must happen outside of the lock because the listener could potentially call
// back into the InputReader's methods, such as getScanCodeState, or become blocked
// on another thread similarly waiting to acquire the InputReader lock thereby
// resulting in a deadlock. This situation is actually quite plausible because the
// listener is actually the input dispatcher, which calls into the window manager,
// which occasionally calls into the input reader.
mQueuedListener->flush();
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void InputReader::processEventsLocked(const RawEvent* rawEvents, size_t count) {
Optimize EventHub to process events in big chunks. When 10 fingers are down, reduces the CPU time spent by the InputReader thread from ~30% to ~5% on Stingray. Change-Id: Icdf7c91cd5d9039ac3beb38ba9021a05e7fabc80
2011-03-18 18:14:26 -07:00
for (const RawEvent* rawEvent = rawEvents; count;) {
int32_t type = rawEvent->type;
size_t batchSize = 1;
if (type < EventHubInterface::FIRST_SYNTHETIC_EVENT) {
int32_t deviceId = rawEvent->deviceId;
while (batchSize < count) {
if (rawEvent[batchSize].type >= EventHubInterface::FIRST_SYNTHETIC_EVENT
|| rawEvent[batchSize].deviceId != deviceId) {
break;
}
batchSize += 1;
}
#if DEBUG_RAW_EVENTS
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("BatchSize: %d Count: %d", batchSize, count);
Optimize EventHub to process events in big chunks. When 10 fingers are down, reduces the CPU time spent by the InputReader thread from ~30% to ~5% on Stingray. Change-Id: Icdf7c91cd5d9039ac3beb38ba9021a05e7fabc80
2011-03-18 18:14:26 -07:00
#endif
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
processEventsForDeviceLocked(deviceId, rawEvent, batchSize);
Optimize EventHub to process events in big chunks. When 10 fingers are down, reduces the CPU time spent by the InputReader thread from ~30% to ~5% on Stingray. Change-Id: Icdf7c91cd5d9039ac3beb38ba9021a05e7fabc80
2011-03-18 18:14:26 -07:00
} else {
switch (rawEvent->type) {
case EventHubInterface::DEVICE_ADDED:
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
addDeviceLocked(rawEvent->when, rawEvent->deviceId);
Optimize EventHub to process events in big chunks. When 10 fingers are down, reduces the CPU time spent by the InputReader thread from ~30% to ~5% on Stingray. Change-Id: Icdf7c91cd5d9039ac3beb38ba9021a05e7fabc80
2011-03-18 18:14:26 -07:00
break;
case EventHubInterface::DEVICE_REMOVED:
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
removeDeviceLocked(rawEvent->when, rawEvent->deviceId);
Optimize EventHub to process events in big chunks. When 10 fingers are down, reduces the CPU time spent by the InputReader thread from ~30% to ~5% on Stingray. Change-Id: Icdf7c91cd5d9039ac3beb38ba9021a05e7fabc80
2011-03-18 18:14:26 -07:00
break;
case EventHubInterface::FINISHED_DEVICE_SCAN:
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
handleConfigurationChangedLocked(rawEvent->when);
Optimize EventHub to process events in big chunks. When 10 fingers are down, reduces the CPU time spent by the InputReader thread from ~30% to ~5% on Stingray. Change-Id: Icdf7c91cd5d9039ac3beb38ba9021a05e7fabc80
2011-03-18 18:14:26 -07:00
break;
default:
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(false); // can't happen
Optimize EventHub to process events in big chunks. When 10 fingers are down, reduces the CPU time spent by the InputReader thread from ~30% to ~5% on Stingray. Change-Id: Icdf7c91cd5d9039ac3beb38ba9021a05e7fabc80
2011-03-18 18:14:26 -07:00
break;
}
}
count -= batchSize;
rawEvent += batchSize;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void InputReader::addDeviceLocked(nsecs_t when, int32_t deviceId) {
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
ssize_t deviceIndex = mDevices.indexOfKey(deviceId);
if (deviceIndex >= 0) {
ALOGW("Ignoring spurious device added event for deviceId %d.", deviceId);
return;
}
Add a unique input device descriptor. The purpose of the input device descriptor is to make it possible to associate persistent settings for each input device, such as the keyboard layout. The descriptor is a hash of the information we have about the device, such as its vendor id, product id, unique id, name, or location. Bug: 6110399 Change-Id: Idb80f946819b3f0dbf4e661bb0a753dbc2b60981
2012-04-06 14:51:01 -07:00
InputDeviceIdentifier identifier = mEventHub->getDeviceIdentifier(deviceId);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
uint32_t classes = mEventHub->getDeviceClasses(deviceId);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Add a unique input device descriptor. The purpose of the input device descriptor is to make it possible to associate persistent settings for each input device, such as the keyboard layout. The descriptor is a hash of the information we have about the device, such as its vendor id, product id, unique id, name, or location. Bug: 6110399 Change-Id: Idb80f946819b3f0dbf4e661bb0a753dbc2b60981
2012-04-06 14:51:01 -07:00
InputDevice* device = createDeviceLocked(deviceId, identifier, classes);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
device->configure(when, &mConfig, 0);
device->reset(when);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
if (device->isIgnored()) {
Add a unique input device descriptor. The purpose of the input device descriptor is to make it possible to associate persistent settings for each input device, such as the keyboard layout. The descriptor is a hash of the information we have about the device, such as its vendor id, product id, unique id, name, or location. Bug: 6110399 Change-Id: Idb80f946819b3f0dbf4e661bb0a753dbc2b60981
2012-04-06 14:51:01 -07:00
ALOGI("Device added: id=%d, name='%s' (ignored non-input device)", deviceId,
identifier.name.string());
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
} else {
Add a unique input device descriptor. The purpose of the input device descriptor is to make it possible to associate persistent settings for each input device, such as the keyboard layout. The descriptor is a hash of the information we have about the device, such as its vendor id, product id, unique id, name, or location. Bug: 6110399 Change-Id: Idb80f946819b3f0dbf4e661bb0a753dbc2b60981
2012-04-06 14:51:01 -07:00
ALOGI("Device added: id=%d, name='%s', sources=0x%08x", deviceId,
identifier.name.string(), device->getSources());
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
mDevices.add(deviceId, device);
bumpGenerationLocked();
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void InputReader::removeDeviceLocked(nsecs_t when, int32_t deviceId) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
InputDevice* device = NULL;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
ssize_t deviceIndex = mDevices.indexOfKey(deviceId);
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
if (deviceIndex < 0) {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW("Ignoring spurious device removed event for deviceId %d.", deviceId);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
return;
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
device = mDevices.valueAt(deviceIndex);
mDevices.removeItemsAt(deviceIndex, 1);
bumpGenerationLocked();
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
if (device->isIgnored()) {
Rename (IF_)LOGI(_IF) to (IF_)ALOGI(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156801 Bug: 5449033 Change-Id: Ib08fe86d23db91ee153e9f91a99a35c42b9208ea
2012-01-04 20:05:49 +00:00
ALOGI("Device removed: id=%d, name='%s' (ignored non-input device)",
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
device->getId(), device->getName().string());
} else {
Rename (IF_)LOGI(_IF) to (IF_)ALOGI(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156801 Bug: 5449033 Change-Id: Ib08fe86d23db91ee153e9f91a99a35c42b9208ea
2012-01-04 20:05:49 +00:00
ALOGI("Device removed: id=%d, name='%s', sources=0x%08x",
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
device->getId(), device->getName().string(), device->getSources());
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
device->reset(when);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
delete device;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
InputDevice* InputReader::createDeviceLocked(int32_t deviceId,
Add a unique input device descriptor. The purpose of the input device descriptor is to make it possible to associate persistent settings for each input device, such as the keyboard layout. The descriptor is a hash of the information we have about the device, such as its vendor id, product id, unique id, name, or location. Bug: 6110399 Change-Id: Idb80f946819b3f0dbf4e661bb0a753dbc2b60981
2012-04-06 14:51:01 -07:00
const InputDeviceIdentifier& identifier, uint32_t classes) {
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
InputDevice* device = new InputDevice(&mContext, deviceId, bumpGenerationLocked(),
identifier, classes);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Wake screen from external HID peripherals. Added some plumbing to enable the policy to intercept motion events when the screen is off to handle wakeup if needed. Added a basic concept of an external device to limit the scope of the wakeup policy to external devices only. The wakeup policy for internal devices should be based on explicit rules such as policy flags in key layout files. Moved isTouchEvent to native. Ensure the dispatcher sends the right event type to userActivity for non-touch pointer events like HOVER_MOVE and SCROLL. Bug: 3193114 Change-Id: I15dbd48a16810dfaf226ff7ad117d46908ca4f86
2011-03-02 19:23:13 -08:00
// External devices.
if (classes & INPUT_DEVICE_CLASS_EXTERNAL) {
device->setExternal(true);
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// Switch-like devices.
if (classes & INPUT_DEVICE_CLASS_SWITCH) {
device->addMapper(new SwitchInputMapper(device));
Add initial gamepad support. Change-Id: I0439648f6eb5405f200e4223c915eb3a418b32b9
2010-06-30 16:10:35 -07:00
}
Add support for input devices that have vibrators. Added a getVibrator() method to InputDevice which returns a Vibrator associated with that input device. Its uses the same API as the system vibrator which makes it easy for applications to be modified to use one or the other. Bug: 6334179 Change-Id: Ifc7f13dbcb778670f3f1c07ccc562334e6109d2e
2012-04-13 04:09:27 -07:00
// Vibrator-like devices.
if (classes & INPUT_DEVICE_CLASS_VIBRATOR) {
device->addMapper(new VibratorInputMapper(device));
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// Keyboard-like devices.
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
uint32_t keyboardSource = 0;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t keyboardType = AINPUT_KEYBOARD_TYPE_NON_ALPHABETIC;
if (classes & INPUT_DEVICE_CLASS_KEYBOARD) {
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
keyboardSource |= AINPUT_SOURCE_KEYBOARD;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
if (classes & INPUT_DEVICE_CLASS_ALPHAKEY) {
keyboardType = AINPUT_KEYBOARD_TYPE_ALPHABETIC;
}
if (classes & INPUT_DEVICE_CLASS_DPAD) {
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
keyboardSource |= AINPUT_SOURCE_DPAD;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
if (classes & INPUT_DEVICE_CLASS_GAMEPAD) {
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
keyboardSource |= AINPUT_SOURCE_GAMEPAD;
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
}
Add initial gamepad support. Change-Id: I0439648f6eb5405f200e4223c915eb3a418b32b9
2010-06-30 16:10:35 -07:00
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
if (keyboardSource != 0) {
device->addMapper(new KeyboardInputMapper(device, keyboardSource, keyboardType));
Add initial gamepad support. Change-Id: I0439648f6eb5405f200e4223c915eb3a418b32b9
2010-06-30 16:10:35 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
// Cursor-like devices.
if (classes & INPUT_DEVICE_CLASS_CURSOR) {
device->addMapper(new CursorInputMapper(device));
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Fix crashes caused by some input devices. The touch screen sometimes reports more than 10 pointers even though that's all we asked for. When this happens, we start dropping events with more than 10 pointers. This confuses applications and causes them to crash. Raised the limit to 16 pointers. Bug: 3331247 The default behavior was to identify all touch devices as touch screens. External devices that are plugged in are more likely to be touch pads not attached to a screen. Changed the default to be a touch pad and renamed some internal constants to avoid confusion. A certain mouse happens to also behave like a touch pad. That caused problems because we would see multiple concurrent traces of motion events coming from the same input device so we would batch them up. Added code to ensure that we don't batch events unless they come from the same *source* in addition to coming from the same *device*. Due to batching or misbehaving drivers, it's possible for the set of pointer ids to be different from what we expect when it comes time to split motion events across windows. As a result, we can generate motion events with 0 pointers. When we try to deliver those events, we cause an error in the InputTransport so we tear down the InputChannel and kill the application. Added code to check out assumption about pointer ids and drop the event gracefully instead. Patched up the tests to take into account the change in default behavior for identifying touch screens and touch pads. Change-Id: Ic364bd4cb4cc6335d4a1213a26d6bdadc7e33505
2011-01-25 16:02:22 -08:00
// Touchscreens and touchpad devices.
if (classes & INPUT_DEVICE_CLASS_TOUCH_MT) {
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
device->addMapper(new MultiTouchInputMapper(device));
Fix crashes caused by some input devices. The touch screen sometimes reports more than 10 pointers even though that's all we asked for. When this happens, we start dropping events with more than 10 pointers. This confuses applications and causes them to crash. Raised the limit to 16 pointers. Bug: 3331247 The default behavior was to identify all touch devices as touch screens. External devices that are plugged in are more likely to be touch pads not attached to a screen. Changed the default to be a touch pad and renamed some internal constants to avoid confusion. A certain mouse happens to also behave like a touch pad. That caused problems because we would see multiple concurrent traces of motion events coming from the same input device so we would batch them up. Added code to ensure that we don't batch events unless they come from the same *source* in addition to coming from the same *device*. Due to batching or misbehaving drivers, it's possible for the set of pointer ids to be different from what we expect when it comes time to split motion events across windows. As a result, we can generate motion events with 0 pointers. When we try to deliver those events, we cause an error in the InputTransport so we tear down the InputChannel and kill the application. Added code to check out assumption about pointer ids and drop the event gracefully instead. Patched up the tests to take into account the change in default behavior for identifying touch screens and touch pads. Change-Id: Ic364bd4cb4cc6335d4a1213a26d6bdadc7e33505
2011-01-25 16:02:22 -08:00
} else if (classes & INPUT_DEVICE_CLASS_TOUCH) {
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
device->addMapper(new SingleTouchInputMapper(device));
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
// Joystick-like devices.
if (classes & INPUT_DEVICE_CLASS_JOYSTICK) {
device->addMapper(new JoystickInputMapper(device));
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
return device;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void InputReader::processEventsForDeviceLocked(int32_t deviceId,
Optimize EventHub to process events in big chunks. When 10 fingers are down, reduces the CPU time spent by the InputReader thread from ~30% to ~5% on Stingray. Change-Id: Icdf7c91cd5d9039ac3beb38ba9021a05e7fabc80
2011-03-18 18:14:26 -07:00
const RawEvent* rawEvents, size_t count) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
ssize_t deviceIndex = mDevices.indexOfKey(deviceId);
if (deviceIndex < 0) {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW("Discarding event for unknown deviceId %d.", deviceId);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
return;
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
InputDevice* device = mDevices.valueAt(deviceIndex);
if (device->isIgnored()) {
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
//ALOGD("Discarding event for ignored deviceId %d.", deviceId);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
return;
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
device->process(rawEvents, count);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void InputReader::timeoutExpiredLocked(nsecs_t when) {
for (size_t i = 0; i < mDevices.size(); i++) {
InputDevice* device = mDevices.valueAt(i);
if (!device->isIgnored()) {
device->timeoutExpired(when);
Refactor how timeouts are calculated. Added a timeout mechanism to EventHub and InputReader so that InputMappers can request timeouts to perform delayed processing of input when needed. Change-Id: Iec2045baaf4e67690b15eef3c09a58d5cac76897
2011-03-17 01:34:19 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Refactor how timeouts are calculated. Added a timeout mechanism to EventHub and InputReader so that InputMappers can request timeouts to perform delayed processing of input when needed. Change-Id: Iec2045baaf4e67690b15eef3c09a58d5cac76897
2011-03-17 01:34:19 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void InputReader::handleConfigurationChangedLocked(nsecs_t when) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// Reset global meta state because it depends on the list of all configured devices.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
updateGlobalMetaStateLocked();
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// Enqueue configuration changed.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
NotifyConfigurationChangedArgs args(when);
mQueuedListener->notifyConfigurationChanged(&args);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void InputReader::refreshConfigurationLocked(uint32_t changes) {
Add a preference panel for mouse speed. Bug: 4124987 Change-Id: I3ce175d268a1d043cf5878481261b1049a15a149
2011-06-02 01:26:32 -07:00
mPolicy->getReaderConfiguration(&mConfig);
mEventHub->setExcludedDevices(mConfig.excludedDeviceNames);
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (changes) {
Rename (IF_)LOGI(_IF) to (IF_)ALOGI(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156801 Bug: 5449033 Change-Id: Ib08fe86d23db91ee153e9f91a99a35c42b9208ea
2012-01-04 20:05:49 +00:00
ALOGI("Reconfiguring input devices. changes=0x%08x", changes);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
nsecs_t now = systemTime(SYSTEM_TIME_MONOTONIC);
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (changes & InputReaderConfiguration::CHANGE_MUST_REOPEN) {
mEventHub->requestReopenDevices();
} else {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
for (size_t i = 0; i < mDevices.size(); i++) {
InputDevice* device = mDevices.valueAt(i);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
device->configure(now, &mConfig, changes);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void InputReader::updateGlobalMetaStateLocked() {
mGlobalMetaState = 0;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
for (size_t i = 0; i < mDevices.size(); i++) {
InputDevice* device = mDevices.valueAt(i);
mGlobalMetaState |= device->getMetaState();
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
int32_t InputReader::getGlobalMetaStateLocked() {
return mGlobalMetaState;
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void InputReader::disableVirtualKeysUntilLocked(nsecs_t time) {
Filter virtual keys after touches. Adds a new virtualKeyQuietTimeMillis configuration resource that sets the duration for which virtual keys will be dropped after recent touches on screen. The default value is 0; it is intended to be overridden per device using a resource overlay. This change is designed to help in two cases: 1. Swipes from touchscreen into virtual key area. 2. Accidental taps in virtual key area while using on-screen keyboard. Bug: 3089163 Change-Id: Ib912d4f8a4df9966a39cd537d3ec7c24afab7225
2011-01-18 15:10:10 -08:00
mDisableVirtualKeysTimeout = time;
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
bool InputReader::shouldDropVirtualKeyLocked(nsecs_t now,
Filter virtual keys after touches. Adds a new virtualKeyQuietTimeMillis configuration resource that sets the duration for which virtual keys will be dropped after recent touches on screen. The default value is 0; it is intended to be overridden per device using a resource overlay. This change is designed to help in two cases: 1. Swipes from touchscreen into virtual key area. 2. Accidental taps in virtual key area while using on-screen keyboard. Bug: 3089163 Change-Id: Ib912d4f8a4df9966a39cd537d3ec7c24afab7225
2011-01-18 15:10:10 -08:00
InputDevice* device, int32_t keyCode, int32_t scanCode) {
if (now < mDisableVirtualKeysTimeout) {
Rename (IF_)LOGI(_IF) to (IF_)ALOGI(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156801 Bug: 5449033 Change-Id: Ib08fe86d23db91ee153e9f91a99a35c42b9208ea
2012-01-04 20:05:49 +00:00
ALOGI("Dropping virtual key from device %s because virtual keys are "
Filter virtual keys after touches. Adds a new virtualKeyQuietTimeMillis configuration resource that sets the duration for which virtual keys will be dropped after recent touches on screen. The default value is 0; it is intended to be overridden per device using a resource overlay. This change is designed to help in two cases: 1. Swipes from touchscreen into virtual key area. 2. Accidental taps in virtual key area while using on-screen keyboard. Bug: 3089163 Change-Id: Ib912d4f8a4df9966a39cd537d3ec7c24afab7225
2011-01-18 15:10:10 -08:00
"temporarily disabled for the next %0.3fms. keyCode=%d, scanCode=%d",
device->getName().string(),
(mDisableVirtualKeysTimeout - now) * 0.000001,
keyCode, scanCode);
return true;
} else {
return false;
}
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void InputReader::fadePointerLocked() {
for (size_t i = 0; i < mDevices.size(); i++) {
InputDevice* device = mDevices.valueAt(i);
device->fadePointer();
}
Fade out the mouse pointer after inactivity or other events. Fades out the mouse pointer: - after 15 seconds of inactivity normally - after 3 seconds of inactivity in lights out mode - after a non-modifier key down - after a touch down Extended the native Looper to support enqueuing time delayed messages. This is used by the PointerController to control pointer fade timing. Change-Id: I87792fea7dbe2d9376c78cf354fe3189a484d9da
2011-03-02 14:41:58 -08:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void InputReader::requestTimeoutAtTimeLocked(nsecs_t when) {
Refactor how timeouts are calculated. Added a timeout mechanism to EventHub and InputReader so that InputMappers can request timeouts to perform delayed processing of input when needed. Change-Id: Iec2045baaf4e67690b15eef3c09a58d5cac76897
2011-03-17 01:34:19 -07:00
if (when < mNextTimeout) {
mNextTimeout = when;
Add support for input devices that have vibrators. Added a getVibrator() method to InputDevice which returns a Vibrator associated with that input device. Its uses the same API as the system vibrator which makes it easy for applications to be modified to use one or the other. Bug: 6334179 Change-Id: Ifc7f13dbcb778670f3f1c07ccc562334e6109d2e
2012-04-13 04:09:27 -07:00
mEventHub->wake();
Refactor how timeouts are calculated. Added a timeout mechanism to EventHub and InputReader so that InputMappers can request timeouts to perform delayed processing of input when needed. Change-Id: Iec2045baaf4e67690b15eef3c09a58d5cac76897
2011-03-17 01:34:19 -07:00
}
}
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
int32_t InputReader::bumpGenerationLocked() {
return ++mGeneration;
}
void InputReader::getInputDevices(Vector<InputDeviceInfo>& outInputDevices) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
AutoMutex _l(mLock);
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
getInputDevicesLocked(outInputDevices);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
void InputReader::getInputDevicesLocked(Vector<InputDeviceInfo>& outInputDevices) {
outInputDevices.clear();
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
size_t numDevices = mDevices.size();
for (size_t i = 0; i < numDevices; i++) {
InputDevice* device = mDevices.valueAt(i);
if (!device->isIgnored()) {
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
outInputDevices.push();
device->getDeviceInfo(&outInputDevices.editTop());
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t InputReader::getKeyCodeState(int32_t deviceId, uint32_t sourceMask,
int32_t keyCode) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
AutoMutex _l(mLock);
return getStateLocked(deviceId, sourceMask, keyCode, &InputDevice::getKeyCodeState);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Add support for new input sources. Added several new coordinate values to MotionEvents to capture touch major/minor area, tool major/minor area and orientation. Renamed NDK input constants per convention. Added InputDevice class in Java which will eventually provide useful information about available input devices. Added APIs for manufacturing new MotionEvent objects with multiple pointers and all necessary coordinate data. Fixed a bug in the input dispatcher where it could get stuck with a pointer down forever. Fixed a bug in the WindowManager where the input window list could end up containing stale removed windows. Fixed a bug in the WindowManager where the input channel was being removed only after the final animation transition had taken place which caused spurious WINDOW DIED log messages to be printed. Change-Id: Ie55084da319b20aad29b28a0499b8dd98bb5da68
2010-07-14 18:48:53 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t InputReader::getScanCodeState(int32_t deviceId, uint32_t sourceMask,
int32_t scanCode) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
AutoMutex _l(mLock);
return getStateLocked(deviceId, sourceMask, scanCode, &InputDevice::getScanCodeState);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Add support for new input sources. Added several new coordinate values to MotionEvents to capture touch major/minor area, tool major/minor area and orientation. Renamed NDK input constants per convention. Added InputDevice class in Java which will eventually provide useful information about available input devices. Added APIs for manufacturing new MotionEvent objects with multiple pointers and all necessary coordinate data. Fixed a bug in the input dispatcher where it could get stuck with a pointer down forever. Fixed a bug in the WindowManager where the input window list could end up containing stale removed windows. Fixed a bug in the WindowManager where the input channel was being removed only after the final animation transition had taken place which caused spurious WINDOW DIED log messages to be printed. Change-Id: Ie55084da319b20aad29b28a0499b8dd98bb5da68
2010-07-14 18:48:53 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t InputReader::getSwitchState(int32_t deviceId, uint32_t sourceMask, int32_t switchCode) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
AutoMutex _l(mLock);
return getStateLocked(deviceId, sourceMask, switchCode, &InputDevice::getSwitchState);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Add support for new input sources. Added several new coordinate values to MotionEvents to capture touch major/minor area, tool major/minor area and orientation. Renamed NDK input constants per convention. Added InputDevice class in Java which will eventually provide useful information about available input devices. Added APIs for manufacturing new MotionEvent objects with multiple pointers and all necessary coordinate data. Fixed a bug in the input dispatcher where it could get stuck with a pointer down forever. Fixed a bug in the WindowManager where the input window list could end up containing stale removed windows. Fixed a bug in the WindowManager where the input channel was being removed only after the final animation transition had taken place which caused spurious WINDOW DIED log messages to be printed. Change-Id: Ie55084da319b20aad29b28a0499b8dd98bb5da68
2010-07-14 18:48:53 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
int32_t InputReader::getStateLocked(int32_t deviceId, uint32_t sourceMask, int32_t code,
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
GetStateFunc getStateFunc) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
int32_t result = AKEY_STATE_UNKNOWN;
if (deviceId >= 0) {
ssize_t deviceIndex = mDevices.indexOfKey(deviceId);
if (deviceIndex >= 0) {
InputDevice* device = mDevices.valueAt(deviceIndex);
if (! device->isIgnored() && sourcesMatchMask(device->getSources(), sourceMask)) {
result = (device->*getStateFunc)(sourceMask, code);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
} else {
size_t numDevices = mDevices.size();
for (size_t i = 0; i < numDevices; i++) {
InputDevice* device = mDevices.valueAt(i);
if (! device->isIgnored() && sourcesMatchMask(device->getSources(), sourceMask)) {
Modify getState to prefer AKEY_STATE_DOWN and AKEY_STATE_UP over AKEY_STATE_UNKNOWN. For the getState() routines that operate on multiple devices / mappers: Return AKEY_STATE_DOWN (or AKEY_STATE_VIRTUAL) if any of the devices/mappers have the key down. Otherwise, return AKEY_STATE_UP if at least one device/mapper returns AKEY_STATE_UP. Otherwise, return AKEY_STATE_UNKNOWN. Change-Id: I1ef6251d73af916b51f408a41c5e4adf5bdc237a
2011-11-14 14:50:45 -08:00
// If any device reports AKEY_STATE_DOWN or AKEY_STATE_VIRTUAL, return that
// value. Otherwise, return AKEY_STATE_UP as long as one device reports it.
int32_t currentResult = (device->*getStateFunc)(sourceMask, code);
if (currentResult >= AKEY_STATE_DOWN) {
return currentResult;
} else if (currentResult == AKEY_STATE_UP) {
result = currentResult;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
}
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
return result;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
bool InputReader::hasKeys(int32_t deviceId, uint32_t sourceMask,
size_t numCodes, const int32_t* keyCodes, uint8_t* outFlags) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
AutoMutex _l(mLock);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
memset(outFlags, 0, numCodes);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
return markSupportedKeyCodesLocked(deviceId, sourceMask, numCodes, keyCodes, outFlags);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
bool InputReader::markSupportedKeyCodesLocked(int32_t deviceId, uint32_t sourceMask,
size_t numCodes, const int32_t* keyCodes, uint8_t* outFlags) {
bool result = false;
if (deviceId >= 0) {
ssize_t deviceIndex = mDevices.indexOfKey(deviceId);
if (deviceIndex >= 0) {
InputDevice* device = mDevices.valueAt(deviceIndex);
if (! device->isIgnored() && sourcesMatchMask(device->getSources(), sourceMask)) {
result = device->markSupportedKeyCodes(sourceMask,
numCodes, keyCodes, outFlags);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
} else {
size_t numDevices = mDevices.size();
for (size_t i = 0; i < numDevices; i++) {
InputDevice* device = mDevices.valueAt(i);
if (! device->isIgnored() && sourcesMatchMask(device->getSources(), sourceMask)) {
result |= device->markSupportedKeyCodes(sourceMask,
numCodes, keyCodes, outFlags);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
return result;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
void InputReader::requestRefreshConfiguration(uint32_t changes) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
AutoMutex _l(mLock);
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (changes) {
bool needWake = !mConfigurationChangesToRefresh;
mConfigurationChangesToRefresh |= changes;
EventHub code cleanup. Use epoll_wait() instead of poll(). Dropped all support for non-Linux platforms. Added a wake-up protocol so that the InputReader can wake up the event loop immediately as needed. Change-Id: Ibf84337bcceb3c2df068c5c637de42a319786d66
2011-06-14 17:09:25 -07:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (needWake) {
mEventHub->wake();
}
}
Add a preference panel for mouse speed. Bug: 4124987 Change-Id: I3ce175d268a1d043cf5878481261b1049a15a149
2011-06-02 01:26:32 -07:00
}
Add support for input devices that have vibrators. Added a getVibrator() method to InputDevice which returns a Vibrator associated with that input device. Its uses the same API as the system vibrator which makes it easy for applications to be modified to use one or the other. Bug: 6334179 Change-Id: Ifc7f13dbcb778670f3f1c07ccc562334e6109d2e
2012-04-13 04:09:27 -07:00
void InputReader::vibrate(int32_t deviceId, const nsecs_t* pattern, size_t patternSize,
ssize_t repeat, int32_t token) {
AutoMutex _l(mLock);
ssize_t deviceIndex = mDevices.indexOfKey(deviceId);
if (deviceIndex >= 0) {
InputDevice* device = mDevices.valueAt(deviceIndex);
device->vibrate(pattern, patternSize, repeat, token);
}
}
void InputReader::cancelVibrate(int32_t deviceId, int32_t token) {
AutoMutex _l(mLock);
ssize_t deviceIndex = mDevices.indexOfKey(deviceId);
if (deviceIndex >= 0) {
InputDevice* device = mDevices.valueAt(deviceIndex);
device->cancelVibrate(token);
}
}
Input dispatcher ANR handling enhancements. This change is essentially a rewrite of the main input dispatcher loop with the target identification folded in. Since the input dispatcher now has all of the window state, it can make better decisions about when to ANR. Added a .5 second deadline for processing app switch keys. This behavior predates Gingerbread but had not previously been ported. Fixed some timing inaccuracies in the ANR accounting that could cause applications to ANR sooner than they should have. Added a mechanism for tracking key and motion events that have been dispatched to a window so that appropriate cancelation events can be synthesized when recovering from ANR. This change helps to keep applications in sync so they don't end up with stuck buttons upon recovery from ANRs. Added more comments to describe the tricky parts of PollLoop. Change-Id: I13dffca27acb436fc383980db536abc4d8b9e6f1
2010-09-08 11:49:43 -07:00
void InputReader::dump(String8& dump) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
AutoMutex _l(mLock);
Minor logging changes to input dispatcher to help with debugging. Added dumpsys reporting to EventHub. Made the formatting a bit clearer. Added 'Locked' suffix to some internal methods of EventHub. Change-Id: Ic449560bcce378f6361895d27c66854e9724abb0
2010-10-01 17:46:21 -07:00
mEventHub->dump(dump);
dump.append("\n");
dump.append("Input Reader State:\n");
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
for (size_t i = 0; i < mDevices.size(); i++) {
mDevices.valueAt(i)->dump(dump);
}
Use ViewConfiguration to seed input system configuration. Fix bug where the pointer presentation would be updated on any input reader timeout rather than only when a pointer gesture is in progress. Bug: 4124987 Change-Id: Ie9bba4a0b3228d55e45e65fa2ede5cd6ba887a08
2011-05-26 19:17:02 -07:00
dump.append(INDENT "Configuration:\n");
dump.append(INDENT2 "ExcludedDeviceNames: [");
for (size_t i = 0; i < mConfig.excludedDeviceNames.size(); i++) {
if (i != 0) {
dump.append(", ");
}
dump.append(mConfig.excludedDeviceNames.itemAt(i).string());
}
dump.append("]\n");
dump.appendFormat(INDENT2 "VirtualKeyQuietTime: %0.1fms\n",
mConfig.virtualKeyQuietTime * 0.000001f);
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
dump.appendFormat(INDENT2 "PointerVelocityControlParameters: "
"scale=%0.3f, lowThreshold=%0.3f, highThreshold=%0.3f, acceleration=%0.3f\n",
mConfig.pointerVelocityControlParameters.scale,
mConfig.pointerVelocityControlParameters.lowThreshold,
mConfig.pointerVelocityControlParameters.highThreshold,
mConfig.pointerVelocityControlParameters.acceleration);
dump.appendFormat(INDENT2 "WheelVelocityControlParameters: "
"scale=%0.3f, lowThreshold=%0.3f, highThreshold=%0.3f, acceleration=%0.3f\n",
mConfig.wheelVelocityControlParameters.scale,
mConfig.wheelVelocityControlParameters.lowThreshold,
mConfig.wheelVelocityControlParameters.highThreshold,
mConfig.wheelVelocityControlParameters.acceleration);
Use ViewConfiguration to seed input system configuration. Fix bug where the pointer presentation would be updated on any input reader timeout rather than only when a pointer gesture is in progress. Bug: 4124987 Change-Id: Ie9bba4a0b3228d55e45e65fa2ede5cd6ba887a08
2011-05-26 19:17:02 -07:00
dump.appendFormat(INDENT2 "PointerGesture:\n");
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
dump.appendFormat(INDENT3 "Enabled: %s\n",
toString(mConfig.pointerGesturesEnabled));
Use ViewConfiguration to seed input system configuration. Fix bug where the pointer presentation would be updated on any input reader timeout rather than only when a pointer gesture is in progress. Bug: 4124987 Change-Id: Ie9bba4a0b3228d55e45e65fa2ede5cd6ba887a08
2011-05-26 19:17:02 -07:00
dump.appendFormat(INDENT3 "QuietInterval: %0.1fms\n",
mConfig.pointerGestureQuietInterval * 0.000001f);
dump.appendFormat(INDENT3 "DragMinSwitchSpeed: %0.1fpx/s\n",
mConfig.pointerGestureDragMinSwitchSpeed);
dump.appendFormat(INDENT3 "TapInterval: %0.1fms\n",
mConfig.pointerGestureTapInterval * 0.000001f);
dump.appendFormat(INDENT3 "TapDragInterval: %0.1fms\n",
mConfig.pointerGestureTapDragInterval * 0.000001f);
dump.appendFormat(INDENT3 "TapSlop: %0.1fpx\n",
mConfig.pointerGestureTapSlop);
dump.appendFormat(INDENT3 "MultitouchSettleInterval: %0.1fms\n",
mConfig.pointerGestureMultitouchSettleInterval * 0.000001f);
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
dump.appendFormat(INDENT3 "MultitouchMinDistance: %0.1fpx\n",
mConfig.pointerGestureMultitouchMinDistance);
Use ViewConfiguration to seed input system configuration. Fix bug where the pointer presentation would be updated on any input reader timeout rather than only when a pointer gesture is in progress. Bug: 4124987 Change-Id: Ie9bba4a0b3228d55e45e65fa2ede5cd6ba887a08
2011-05-26 19:17:02 -07:00
dump.appendFormat(INDENT3 "SwipeTransitionAngleCosine: %0.1f\n",
mConfig.pointerGestureSwipeTransitionAngleCosine);
dump.appendFormat(INDENT3 "SwipeMaxWidthRatio: %0.1f\n",
mConfig.pointerGestureSwipeMaxWidthRatio);
dump.appendFormat(INDENT3 "MovementSpeedRatio: %0.1f\n",
mConfig.pointerGestureMovementSpeedRatio);
dump.appendFormat(INDENT3 "ZoomSpeedRatio: %0.1f\n",
mConfig.pointerGestureZoomSpeedRatio);
Input dispatcher ANR handling enhancements. This change is essentially a rewrite of the main input dispatcher loop with the target identification folded in. Since the input dispatcher now has all of the window state, it can make better decisions about when to ANR. Added a .5 second deadline for processing app switch keys. This behavior predates Gingerbread but had not previously been ported. Fixed some timing inaccuracies in the ANR accounting that could cause applications to ANR sooner than they should have. Added a mechanism for tracking key and motion events that have been dispatched to a window so that appropriate cancelation events can be synthesized when recovering from ANR. This change helps to keep applications in sync so they don't end up with stuck buttons upon recovery from ANRs. Added more comments to describe the tricky parts of PollLoop. Change-Id: I13dffca27acb436fc383980db536abc4d8b9e6f1
2010-09-08 11:49:43 -07:00
}
Add input system to Watchdog. Bug: 5094994 Change-Id: I153866958efc64ac19bda8b997c1c9f6ad425ec4
2011-08-10 16:25:21 -07:00
void InputReader::monitor() {
// Acquire and release the lock to ensure that the reader has not deadlocked.
mLock.lock();
Improve watchdog monitor for InputReader and InputDispatcher. Try harder to test for liveness. There are situations where the lock might not be held but the input system is stuck in a callback into the window manager policy that has hung. Bug: 5094994 Change-Id: Iff88655512a5dc8bbb4615be65f4115e975c020b
2012-01-27 17:32:06 -08:00
mEventHub->wake();
mReaderIsAliveCondition.wait(mLock);
Add input system to Watchdog. Bug: 5094994 Change-Id: I153866958efc64ac19bda8b997c1c9f6ad425ec4
2011-08-10 16:25:21 -07:00
mLock.unlock();
// Check the EventHub
mEventHub->monitor();
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// --- InputReader::ContextImpl ---
InputReader::ContextImpl::ContextImpl(InputReader* reader) :
mReader(reader) {
}
void InputReader::ContextImpl::updateGlobalMetaState() {
// lock is already held by the input loop
mReader->updateGlobalMetaStateLocked();
}
int32_t InputReader::ContextImpl::getGlobalMetaState() {
// lock is already held by the input loop
return mReader->getGlobalMetaStateLocked();
}
void InputReader::ContextImpl::disableVirtualKeysUntil(nsecs_t time) {
// lock is already held by the input loop
mReader->disableVirtualKeysUntilLocked(time);
}
bool InputReader::ContextImpl::shouldDropVirtualKey(nsecs_t now,
InputDevice* device, int32_t keyCode, int32_t scanCode) {
// lock is already held by the input loop
return mReader->shouldDropVirtualKeyLocked(now, device, keyCode, scanCode);
}
void InputReader::ContextImpl::fadePointer() {
// lock is already held by the input loop
mReader->fadePointerLocked();
}
void InputReader::ContextImpl::requestTimeoutAtTime(nsecs_t when) {
// lock is already held by the input loop
mReader->requestTimeoutAtTimeLocked(when);
}
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
int32_t InputReader::ContextImpl::bumpGeneration() {
// lock is already held by the input loop
return mReader->bumpGenerationLocked();
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
InputReaderPolicyInterface* InputReader::ContextImpl::getPolicy() {
return mReader->mPolicy.get();
}
InputListenerInterface* InputReader::ContextImpl::getListener() {
return mReader->mQueuedListener.get();
}
EventHubInterface* InputReader::ContextImpl::getEventHub() {
return mReader->mEventHub.get();
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// --- InputReaderThread ---
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
InputReaderThread::InputReaderThread(const sp<InputReaderInterface>& reader) :
Thread(/*canCallJava*/ true), mReader(reader) {
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
InputReaderThread::~InputReaderThread() {
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
bool InputReaderThread::threadLoop() {
mReader->loopOnce();
return true;
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// --- InputDevice ---
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
InputDevice::InputDevice(InputReaderContext* context, int32_t id, int32_t generation,
Add a unique input device descriptor. The purpose of the input device descriptor is to make it possible to associate persistent settings for each input device, such as the keyboard layout. The descriptor is a hash of the information we have about the device, such as its vendor id, product id, unique id, name, or location. Bug: 6110399 Change-Id: Idb80f946819b3f0dbf4e661bb0a753dbc2b60981
2012-04-06 14:51:01 -07:00
const InputDeviceIdentifier& identifier, uint32_t classes) :
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
mContext(context), mId(id), mGeneration(generation),
mIdentifier(identifier), mClasses(classes),
Support composite touch / joystick devices better. This change enables the joystick input mapper to handle any axes that are not claimed by the touch input mapper, which makes auxiliary controls such as wheels / knobs accessible. Change-Id: I01ee7f342ac91acfcb4ccb6676fd52b3d5bf31a0
2011-08-31 12:56:34 -07:00
mSources(0), mIsExternal(false), mDropUntilNextSync(false) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
InputDevice::~InputDevice() {
size_t numMappers = mMappers.size();
for (size_t i = 0; i < numMappers; i++) {
delete mMappers[i];
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
mMappers.clear();
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
void InputDevice::dump(String8& dump) {
InputDeviceInfo deviceInfo;
getDeviceInfo(& deviceInfo);
Improve support for external keyboards. Use Vendor ID, Product ID and optionally the Version to locate keymaps and configuration files for external devices. Moved virtual key definition parsing to native code so that EventHub can identify touch screens with virtual keys and load the appropriate key layout file. Cleaned up a lot of old code in EventHub. Fixed a regression in ViewRoot's fallback event handling. Fixed a minor bug in FileMap that caused it to try to munmap or close invalid handled when released if the attempt to map the file failed. Added a couple of new String8 conveniences for formatting strings. Modified Tokenizer to fall back to open+read when mmap fails since we can't mmap sysfs files as needed to open the virtual key definition files in /sys/board_properties/. Change-Id: I6ca5e5f9547619fd082ddac47e87ce185da69ee6
2010-12-02 13:50:46 -08:00
dump.appendFormat(INDENT "Device %d: %s\n", deviceInfo.getId(),
Get alias for Bluetooth devices. Bluetooth devices can be renamed by the user. Make the input system aware of the user-specified name and transparently pass it down to applications. This enables the keyboard layout picker Settings UI to use device names that are consistent with what the user set in the Bluetooth UI. Bug: 6363157 Change-Id: I8eea26ce2c69c2a3f09c8de02e9e847610e0419c
2012-04-20 19:28:00 -07:00
deviceInfo.getDisplayName().string());
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
dump.appendFormat(INDENT2 "Generation: %d\n", mGeneration);
Wake screen from external HID peripherals. Added some plumbing to enable the policy to intercept motion events when the screen is off to handle wakeup if needed. Added a basic concept of an external device to limit the scope of the wakeup policy to external devices only. The wakeup policy for internal devices should be based on explicit rules such as policy flags in key layout files. Moved isTouchEvent to native. Ensure the dispatcher sends the right event type to userActivity for non-touch pointer events like HOVER_MOVE and SCROLL. Bug: 3193114 Change-Id: I15dbd48a16810dfaf226ff7ad117d46908ca4f86
2011-03-02 19:23:13 -08:00
dump.appendFormat(INDENT2 "IsExternal: %s\n", toString(mIsExternal));
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
dump.appendFormat(INDENT2 "Sources: 0x%08x\n", deviceInfo.getSources());
dump.appendFormat(INDENT2 "KeyboardType: %d\n", deviceInfo.getKeyboardType());
Add new hover move action and scroll wheel plumbing. Added support for tracking the mouse position even when the mouse button is not pressed. To avoid confusing existing applications, mouse movements are reported using the new ACTION_HOVER_MOVE action when the mouse button is not pressed. Added some more plumbing for the scroll wheel axes. The values are reported to Views but they are not yet handled by the framework. Change-Id: I1706be850d25cf34e5adf880bbed5cc3265cf4b1
2011-02-19 05:07:28 -08:00
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
const Vector<InputDeviceInfo::MotionRange>& ranges = deviceInfo.getMotionRanges();
Add new hover move action and scroll wheel plumbing. Added support for tracking the mouse position even when the mouse button is not pressed. To avoid confusing existing applications, mouse movements are reported using the new ACTION_HOVER_MOVE action when the mouse button is not pressed. Added some more plumbing for the scroll wheel axes. The values are reported to Views but they are not yet handled by the framework. Change-Id: I1706be850d25cf34e5adf880bbed5cc3265cf4b1
2011-02-19 05:07:28 -08:00
if (!ranges.isEmpty()) {
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
dump.append(INDENT2 "Motion Ranges:\n");
Add new hover move action and scroll wheel plumbing. Added support for tracking the mouse position even when the mouse button is not pressed. To avoid confusing existing applications, mouse movements are reported using the new ACTION_HOVER_MOVE action when the mouse button is not pressed. Added some more plumbing for the scroll wheel axes. The values are reported to Views but they are not yet handled by the framework. Change-Id: I1706be850d25cf34e5adf880bbed5cc3265cf4b1
2011-02-19 05:07:28 -08:00
for (size_t i = 0; i < ranges.size(); i++) {
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
const InputDeviceInfo::MotionRange& range = ranges.itemAt(i);
const char* label = getAxisLabel(range.axis);
Add new hover move action and scroll wheel plumbing. Added support for tracking the mouse position even when the mouse button is not pressed. To avoid confusing existing applications, mouse movements are reported using the new ACTION_HOVER_MOVE action when the mouse button is not pressed. Added some more plumbing for the scroll wheel axes. The values are reported to Views but they are not yet handled by the framework. Change-Id: I1706be850d25cf34e5adf880bbed5cc3265cf4b1
2011-02-19 05:07:28 -08:00
char name[32];
if (label) {
strncpy(name, label, sizeof(name));
name[sizeof(name) - 1] = '\0';
} else {
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
snprintf(name, sizeof(name), "%d", range.axis);
Add new hover move action and scroll wheel plumbing. Added support for tracking the mouse position even when the mouse button is not pressed. To avoid confusing existing applications, mouse movements are reported using the new ACTION_HOVER_MOVE action when the mouse button is not pressed. Added some more plumbing for the scroll wheel axes. The values are reported to Views but they are not yet handled by the framework. Change-Id: I1706be850d25cf34e5adf880bbed5cc3265cf4b1
2011-02-19 05:07:28 -08:00
}
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
dump.appendFormat(INDENT3 "%s: source=0x%08x, "
"min=%0.3f, max=%0.3f, flat=%0.3f, fuzz=%0.3f\n",
name, range.source, range.min, range.max, range.flat, range.fuzz);
Add new hover move action and scroll wheel plumbing. Added support for tracking the mouse position even when the mouse button is not pressed. To avoid confusing existing applications, mouse movements are reported using the new ACTION_HOVER_MOVE action when the mouse button is not pressed. Added some more plumbing for the scroll wheel axes. The values are reported to Views but they are not yet handled by the framework. Change-Id: I1706be850d25cf34e5adf880bbed5cc3265cf4b1
2011-02-19 05:07:28 -08:00
}
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
}
size_t numMappers = mMappers.size();
for (size_t i = 0; i < numMappers; i++) {
InputMapper* mapper = mMappers[i];
mapper->dump(dump);
}
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
void InputDevice::addMapper(InputMapper* mapper) {
mMappers.add(mapper);
}
Native input event dispatching. Target identification is now fully native. Fixed a couple of minor issues related to input injection. Native input enabled by default, can be disabled by setting WindowManagerPolicy.ENABLE_NATIVE_INPUT_DISPATCH to false. Change-Id: I7edf66ed3e987cc9306ad4743ac57a116af452ff
2010-06-22 01:27:15 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void InputDevice::configure(nsecs_t when, const InputReaderConfiguration* config, uint32_t changes) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
mSources = 0;
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (!isIgnored()) {
if (!changes) { // first time only
mContext->getEventHub()->getConfiguration(mId, &mConfiguration);
}
Support loading keyboard layout overlays from resources. Added the concept of a keyboard layout overlay, which is a key character map file that has "type OVERLAY". Added support for loading keyboard layout overlays from resources dynamically. The layouts are reloaded whenever they are changed in the Settings application or an application is installed. This is somewhat more aggressive than necessary so we might want to optimize it later. Before system-ready, the input system uses just the generic keyboard layouts that are included on the device system image. After system-ready, it considers the user's selected keyboard layout overlay and attempts to load it as necessary. We need to wait until system-ready before doing this because we need to be in a state where it is safe to start applications or access their resources. Bug: 6110399 Change-Id: Iae0886d3356649b0d2440aa00910a888cedd8323
2012-04-17 16:52:41 -07:00
if (!changes || (changes & InputReaderConfiguration::CHANGE_KEYBOARD_LAYOUTS)) {
Don't load keyboard layout overlay for virtual keyboard. Bug: 6110399 Change-Id: I07c85cf991790ed8766b097acc39295a8065a5ea
2012-04-19 11:14:33 -07:00
if (!(mClasses & INPUT_DEVICE_CLASS_VIRTUAL)) {
sp<KeyCharacterMap> keyboardLayout =
mContext->getPolicy()->getKeyboardLayoutOverlay(mIdentifier.descriptor);
if (mContext->getEventHub()->setKeyboardLayoutOverlay(mId, keyboardLayout)) {
bumpGeneration();
}
Support loading keyboard layout overlays from resources. Added the concept of a keyboard layout overlay, which is a key character map file that has "type OVERLAY". Added support for loading keyboard layout overlays from resources dynamically. The layouts are reloaded whenever they are changed in the Settings application or an application is installed. This is somewhat more aggressive than necessary so we might want to optimize it later. Before system-ready, the input system uses just the generic keyboard layouts that are included on the device system image. After system-ready, it considers the user's selected keyboard layout overlay and attempts to load it as necessary. We need to wait until system-ready before doing this because we need to be in a state where it is safe to start applications or access their resources. Bug: 6110399 Change-Id: Iae0886d3356649b0d2440aa00910a888cedd8323
2012-04-17 16:52:41 -07:00
}
}
Get alias for Bluetooth devices. Bluetooth devices can be renamed by the user. Make the input system aware of the user-specified name and transparently pass it down to applications. This enables the keyboard layout picker Settings UI to use device names that are consistent with what the user set in the Bluetooth UI. Bug: 6363157 Change-Id: I8eea26ce2c69c2a3f09c8de02e9e847610e0419c
2012-04-20 19:28:00 -07:00
if (!changes || (changes & InputReaderConfiguration::CHANGE_DEVICE_ALIAS)) {
if (!(mClasses & INPUT_DEVICE_CLASS_VIRTUAL)) {
String8 alias = mContext->getPolicy()->getDeviceAlias(mIdentifier);
if (mAlias != alias) {
mAlias = alias;
bumpGeneration();
}
}
}
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
size_t numMappers = mMappers.size();
for (size_t i = 0; i < numMappers; i++) {
InputMapper* mapper = mMappers[i];
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mapper->configure(when, config, changes);
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
mSources |= mapper->getSources();
}
Native input event dispatching. Target identification is now fully native. Fixed a couple of minor issues related to input injection. Native input enabled by default, can be disabled by setting WindowManagerPolicy.ENABLE_NATIVE_INPUT_DISPATCH to false. Change-Id: I7edf66ed3e987cc9306ad4743ac57a116af452ff
2010-06-22 01:27:15 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input event dispatching. Target identification is now fully native. Fixed a couple of minor issues related to input injection. Native input enabled by default, can be disabled by setting WindowManagerPolicy.ENABLE_NATIVE_INPUT_DISPATCH to false. Change-Id: I7edf66ed3e987cc9306ad4743ac57a116af452ff
2010-06-22 01:27:15 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void InputDevice::reset(nsecs_t when) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
size_t numMappers = mMappers.size();
for (size_t i = 0; i < numMappers; i++) {
InputMapper* mapper = mMappers[i];
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mapper->reset(when);
Native input event dispatching. Target identification is now fully native. Fixed a couple of minor issues related to input injection. Native input enabled by default, can be disabled by setting WindowManagerPolicy.ENABLE_NATIVE_INPUT_DISPATCH to false. Change-Id: I7edf66ed3e987cc9306ad4743ac57a116af452ff
2010-06-22 01:27:15 -07:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mContext->updateGlobalMetaState();
notifyReset(when);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input event dispatching. Target identification is now fully native. Fixed a couple of minor issues related to input injection. Native input enabled by default, can be disabled by setting WindowManagerPolicy.ENABLE_NATIVE_INPUT_DISPATCH to false. Change-Id: I7edf66ed3e987cc9306ad4743ac57a116af452ff
2010-06-22 01:27:15 -07:00
Optimize EventHub to process events in big chunks. When 10 fingers are down, reduces the CPU time spent by the InputReader thread from ~30% to ~5% on Stingray. Change-Id: Icdf7c91cd5d9039ac3beb38ba9021a05e7fabc80
2011-03-18 18:14:26 -07:00
void InputDevice::process(const RawEvent* rawEvents, size_t count) {
// Process all of the events in order for each mapper.
// We cannot simply ask each mapper to process them in bulk because mappers may
// have side-effects that must be interleaved. For example, joystick movement events and
// gamepad button presses are handled by different mappers but they should be dispatched
// in the order received.
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
size_t numMappers = mMappers.size();
Optimize EventHub to process events in big chunks. When 10 fingers are down, reduces the CPU time spent by the InputReader thread from ~30% to ~5% on Stingray. Change-Id: Icdf7c91cd5d9039ac3beb38ba9021a05e7fabc80
2011-03-18 18:14:26 -07:00
for (const RawEvent* rawEvent = rawEvents; count--; rawEvent++) {
#if DEBUG_RAW_EVENTS
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
ALOGD("Input event: device=%d type=0x%04x code=0x%04x value=0x%08x",
rawEvent->deviceId, rawEvent->type, rawEvent->code, rawEvent->value);
Optimize EventHub to process events in big chunks. When 10 fingers are down, reduces the CPU time spent by the InputReader thread from ~30% to ~5% on Stingray. Change-Id: Icdf7c91cd5d9039ac3beb38ba9021a05e7fabc80
2011-03-18 18:14:26 -07:00
#endif
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
if (mDropUntilNextSync) {
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
if (rawEvent->type == EV_SYN && rawEvent->code == SYN_REPORT) {
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
mDropUntilNextSync = false;
#if DEBUG_RAW_EVENTS
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Recovered from input event buffer overrun.");
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
#endif
} else {
#if DEBUG_RAW_EVENTS
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Dropped input event while waiting for next input sync.");
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
#endif
}
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
} else if (rawEvent->type == EV_SYN && rawEvent->code == SYN_DROPPED) {
Add a unique input device descriptor. The purpose of the input device descriptor is to make it possible to associate persistent settings for each input device, such as the keyboard layout. The descriptor is a hash of the information we have about the device, such as its vendor id, product id, unique id, name, or location. Bug: 6110399 Change-Id: Idb80f946819b3f0dbf4e661bb0a753dbc2b60981
2012-04-06 14:51:01 -07:00
ALOGI("Detected input event buffer overrun for device %s.", getName().string());
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
mDropUntilNextSync = true;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
reset(rawEvent->when);
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
} else {
for (size_t i = 0; i < numMappers; i++) {
InputMapper* mapper = mMappers[i];
mapper->process(rawEvent);
}
Optimize EventHub to process events in big chunks. When 10 fingers are down, reduces the CPU time spent by the InputReader thread from ~30% to ~5% on Stingray. Change-Id: Icdf7c91cd5d9039ac3beb38ba9021a05e7fabc80
2011-03-18 18:14:26 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor how timeouts are calculated. Added a timeout mechanism to EventHub and InputReader so that InputMappers can request timeouts to perform delayed processing of input when needed. Change-Id: Iec2045baaf4e67690b15eef3c09a58d5cac76897
2011-03-17 01:34:19 -07:00
void InputDevice::timeoutExpired(nsecs_t when) {
size_t numMappers = mMappers.size();
for (size_t i = 0; i < numMappers; i++) {
InputMapper* mapper = mMappers[i];
mapper->timeoutExpired(when);
}
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
void InputDevice::getDeviceInfo(InputDeviceInfo* outDeviceInfo) {
Improve handling of built-in keyboard. The window manager policy made some incorrect assumptions about the meaning of the Configuration.keyboard field. We need to be more careful about distinguishing between built-in and external keyboards. Most of this change is to move the determination of the parts of the Configuration related to input devices into the WindowManagerService leveraging new features of the InputManagerService to good effect. Then we plumb through the flag that indicates whether a device is internal or external so that we can be more particular about how the lid switch effects changes to the Configuration. Bug: 6424373 Change-Id: I36a1c22ade35e578955465a25940a33f227b9763
2012-05-01 15:54:03 -07:00
outDeviceInfo->initialize(mId, mGeneration, mIdentifier, mAlias, mIsExternal);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
size_t numMappers = mMappers.size();
for (size_t i = 0; i < numMappers; i++) {
InputMapper* mapper = mMappers[i];
mapper->populateDeviceInfo(outDeviceInfo);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t InputDevice::getKeyCodeState(uint32_t sourceMask, int32_t keyCode) {
return getState(sourceMask, keyCode, & InputMapper::getKeyCodeState);
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t InputDevice::getScanCodeState(uint32_t sourceMask, int32_t scanCode) {
return getState(sourceMask, scanCode, & InputMapper::getScanCodeState);
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t InputDevice::getSwitchState(uint32_t sourceMask, int32_t switchCode) {
return getState(sourceMask, switchCode, & InputMapper::getSwitchState);
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t InputDevice::getState(uint32_t sourceMask, int32_t code, GetStateFunc getStateFunc) {
int32_t result = AKEY_STATE_UNKNOWN;
size_t numMappers = mMappers.size();
for (size_t i = 0; i < numMappers; i++) {
InputMapper* mapper = mMappers[i];
if (sourcesMatchMask(mapper->getSources(), sourceMask)) {
Modify getState to prefer AKEY_STATE_DOWN and AKEY_STATE_UP over AKEY_STATE_UNKNOWN. For the getState() routines that operate on multiple devices / mappers: Return AKEY_STATE_DOWN (or AKEY_STATE_VIRTUAL) if any of the devices/mappers have the key down. Otherwise, return AKEY_STATE_UP if at least one device/mapper returns AKEY_STATE_UP. Otherwise, return AKEY_STATE_UNKNOWN. Change-Id: I1ef6251d73af916b51f408a41c5e4adf5bdc237a
2011-11-14 14:50:45 -08:00
// If any mapper reports AKEY_STATE_DOWN or AKEY_STATE_VIRTUAL, return that
// value. Otherwise, return AKEY_STATE_UP as long as one mapper reports it.
int32_t currentResult = (mapper->*getStateFunc)(sourceMask, code);
if (currentResult >= AKEY_STATE_DOWN) {
return currentResult;
} else if (currentResult == AKEY_STATE_UP) {
result = currentResult;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
return result;
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
bool InputDevice::markSupportedKeyCodes(uint32_t sourceMask, size_t numCodes,
const int32_t* keyCodes, uint8_t* outFlags) {
bool result = false;
size_t numMappers = mMappers.size();
for (size_t i = 0; i < numMappers; i++) {
InputMapper* mapper = mMappers[i];
if (sourcesMatchMask(mapper->getSources(), sourceMask)) {
result |= mapper->markSupportedKeyCodes(sourceMask, numCodes, keyCodes, outFlags);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
return result;
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Add support for input devices that have vibrators. Added a getVibrator() method to InputDevice which returns a Vibrator associated with that input device. Its uses the same API as the system vibrator which makes it easy for applications to be modified to use one or the other. Bug: 6334179 Change-Id: Ifc7f13dbcb778670f3f1c07ccc562334e6109d2e
2012-04-13 04:09:27 -07:00
void InputDevice::vibrate(const nsecs_t* pattern, size_t patternSize, ssize_t repeat,
int32_t token) {
size_t numMappers = mMappers.size();
for (size_t i = 0; i < numMappers; i++) {
InputMapper* mapper = mMappers[i];
mapper->vibrate(pattern, patternSize, repeat, token);
}
}
void InputDevice::cancelVibrate(int32_t token) {
size_t numMappers = mMappers.size();
for (size_t i = 0; i < numMappers; i++) {
InputMapper* mapper = mMappers[i];
mapper->cancelVibrate(token);
}
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t InputDevice::getMetaState() {
int32_t result = 0;
size_t numMappers = mMappers.size();
for (size_t i = 0; i < numMappers; i++) {
InputMapper* mapper = mMappers[i];
result |= mapper->getMetaState();
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
return result;
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Fade out the mouse pointer after inactivity or other events. Fades out the mouse pointer: - after 15 seconds of inactivity normally - after 3 seconds of inactivity in lights out mode - after a non-modifier key down - after a touch down Extended the native Looper to support enqueuing time delayed messages. This is used by the PointerController to control pointer fade timing. Change-Id: I87792fea7dbe2d9376c78cf354fe3189a484d9da
2011-03-02 14:41:58 -08:00
void InputDevice::fadePointer() {
size_t numMappers = mMappers.size();
for (size_t i = 0; i < numMappers; i++) {
InputMapper* mapper = mMappers[i];
mapper->fadePointer();
}
}
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
void InputDevice::bumpGeneration() {
mGeneration = mContext->bumpGeneration();
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void InputDevice::notifyReset(nsecs_t when) {
NotifyDeviceResetArgs args(when, mId);
mContext->getListener()->notifyDeviceReset(&args);
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
// --- CursorButtonAccumulator ---
CursorButtonAccumulator::CursorButtonAccumulator() {
clearButtons();
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void CursorButtonAccumulator::reset(InputDevice* device) {
mBtnLeft = device->isKeyPressed(BTN_LEFT);
mBtnRight = device->isKeyPressed(BTN_RIGHT);
mBtnMiddle = device->isKeyPressed(BTN_MIDDLE);
mBtnBack = device->isKeyPressed(BTN_BACK);
mBtnSide = device->isKeyPressed(BTN_SIDE);
mBtnForward = device->isKeyPressed(BTN_FORWARD);
mBtnExtra = device->isKeyPressed(BTN_EXTRA);
mBtnTask = device->isKeyPressed(BTN_TASK);
}
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
void CursorButtonAccumulator::clearButtons() {
mBtnLeft = 0;
mBtnRight = 0;
mBtnMiddle = 0;
mBtnBack = 0;
mBtnSide = 0;
mBtnForward = 0;
mBtnExtra = 0;
mBtnTask = 0;
}
void CursorButtonAccumulator::process(const RawEvent* rawEvent) {
if (rawEvent->type == EV_KEY) {
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
switch (rawEvent->code) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
case BTN_LEFT:
mBtnLeft = rawEvent->value;
break;
case BTN_RIGHT:
mBtnRight = rawEvent->value;
break;
case BTN_MIDDLE:
mBtnMiddle = rawEvent->value;
break;
case BTN_BACK:
mBtnBack = rawEvent->value;
break;
case BTN_SIDE:
mBtnSide = rawEvent->value;
break;
case BTN_FORWARD:
mBtnForward = rawEvent->value;
break;
case BTN_EXTRA:
mBtnExtra = rawEvent->value;
break;
case BTN_TASK:
mBtnTask = rawEvent->value;
break;
}
}
}
uint32_t CursorButtonAccumulator::getButtonState() const {
uint32_t result = 0;
if (mBtnLeft) {
result |= AMOTION_EVENT_BUTTON_PRIMARY;
}
if (mBtnRight) {
result |= AMOTION_EVENT_BUTTON_SECONDARY;
}
if (mBtnMiddle) {
result |= AMOTION_EVENT_BUTTON_TERTIARY;
}
if (mBtnBack || mBtnSide) {
result |= AMOTION_EVENT_BUTTON_BACK;
}
if (mBtnForward || mBtnExtra) {
result |= AMOTION_EVENT_BUTTON_FORWARD;
}
return result;
}
// --- CursorMotionAccumulator ---
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
CursorMotionAccumulator::CursorMotionAccumulator() {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
clearRelativeAxes();
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void CursorMotionAccumulator::reset(InputDevice* device) {
clearRelativeAxes();
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
}
void CursorMotionAccumulator::clearRelativeAxes() {
mRelX = 0;
mRelY = 0;
}
void CursorMotionAccumulator::process(const RawEvent* rawEvent) {
if (rawEvent->type == EV_REL) {
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
switch (rawEvent->code) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
case REL_X:
mRelX = rawEvent->value;
break;
case REL_Y:
mRelY = rawEvent->value;
break;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
}
}
}
void CursorMotionAccumulator::finishSync() {
clearRelativeAxes();
}
// --- CursorScrollAccumulator ---
CursorScrollAccumulator::CursorScrollAccumulator() :
mHaveRelWheel(false), mHaveRelHWheel(false) {
clearRelativeAxes();
}
void CursorScrollAccumulator::configure(InputDevice* device) {
mHaveRelWheel = device->getEventHub()->hasRelativeAxis(device->getId(), REL_WHEEL);
mHaveRelHWheel = device->getEventHub()->hasRelativeAxis(device->getId(), REL_HWHEEL);
}
void CursorScrollAccumulator::reset(InputDevice* device) {
clearRelativeAxes();
}
void CursorScrollAccumulator::clearRelativeAxes() {
mRelWheel = 0;
mRelHWheel = 0;
}
void CursorScrollAccumulator::process(const RawEvent* rawEvent) {
if (rawEvent->type == EV_REL) {
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
switch (rawEvent->code) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
case REL_WHEEL:
mRelWheel = rawEvent->value;
break;
case REL_HWHEEL:
mRelHWheel = rawEvent->value;
break;
}
}
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void CursorScrollAccumulator::finishSync() {
clearRelativeAxes();
}
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
// --- TouchButtonAccumulator ---
TouchButtonAccumulator::TouchButtonAccumulator() :
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
mHaveBtnTouch(false), mHaveStylus(false) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
clearButtons();
}
void TouchButtonAccumulator::configure(InputDevice* device) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mHaveBtnTouch = device->hasKey(BTN_TOUCH);
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
mHaveStylus = device->hasKey(BTN_TOOL_PEN)
|| device->hasKey(BTN_TOOL_RUBBER)
|| device->hasKey(BTN_TOOL_BRUSH)
|| device->hasKey(BTN_TOOL_PENCIL)
|| device->hasKey(BTN_TOOL_AIRBRUSH);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
}
void TouchButtonAccumulator::reset(InputDevice* device) {
mBtnTouch = device->isKeyPressed(BTN_TOUCH);
mBtnStylus = device->isKeyPressed(BTN_STYLUS);
mBtnStylus2 = device->isKeyPressed(BTN_STYLUS);
mBtnToolFinger = device->isKeyPressed(BTN_TOOL_FINGER);
mBtnToolPen = device->isKeyPressed(BTN_TOOL_PEN);
mBtnToolRubber = device->isKeyPressed(BTN_TOOL_RUBBER);
mBtnToolBrush = device->isKeyPressed(BTN_TOOL_BRUSH);
mBtnToolPencil = device->isKeyPressed(BTN_TOOL_PENCIL);
mBtnToolAirbrush = device->isKeyPressed(BTN_TOOL_AIRBRUSH);
mBtnToolMouse = device->isKeyPressed(BTN_TOOL_MOUSE);
mBtnToolLens = device->isKeyPressed(BTN_TOOL_LENS);
Support BTN_TOOL_*TAP as synonyms for BTN_TOOL_FINGER. Bug: 5205301 Some drivers report one finger as BTN_TOOL_FINGER, two as BTN_TOOL_DOUBLETAP, three as BTN_TOOL_TRIPLETAP and four as BTN_TOOL_QUADTAP. Since we care about the tool type, we need to handle _DOUBLE/_TRIPLE/_QUAD tap in the same way we handle _FINGER. Change-Id: I8eb83d2a2bada9ac32d07619c7eea84e924316b8
2011-08-23 17:31:25 -07:00
mBtnToolDoubleTap = device->isKeyPressed(BTN_TOOL_DOUBLETAP);
mBtnToolTripleTap = device->isKeyPressed(BTN_TOOL_TRIPLETAP);
mBtnToolQuadTap = device->isKeyPressed(BTN_TOOL_QUADTAP);
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
}
void TouchButtonAccumulator::clearButtons() {
mBtnTouch = 0;
mBtnStylus = 0;
mBtnStylus2 = 0;
mBtnToolFinger = 0;
mBtnToolPen = 0;
mBtnToolRubber = 0;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mBtnToolBrush = 0;
mBtnToolPencil = 0;
mBtnToolAirbrush = 0;
mBtnToolMouse = 0;
mBtnToolLens = 0;
Support BTN_TOOL_*TAP as synonyms for BTN_TOOL_FINGER. Bug: 5205301 Some drivers report one finger as BTN_TOOL_FINGER, two as BTN_TOOL_DOUBLETAP, three as BTN_TOOL_TRIPLETAP and four as BTN_TOOL_QUADTAP. Since we care about the tool type, we need to handle _DOUBLE/_TRIPLE/_QUAD tap in the same way we handle _FINGER. Change-Id: I8eb83d2a2bada9ac32d07619c7eea84e924316b8
2011-08-23 17:31:25 -07:00
mBtnToolDoubleTap = 0;
mBtnToolTripleTap = 0;
mBtnToolQuadTap = 0;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
}
void TouchButtonAccumulator::process(const RawEvent* rawEvent) {
if (rawEvent->type == EV_KEY) {
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
switch (rawEvent->code) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
case BTN_TOUCH:
mBtnTouch = rawEvent->value;
break;
case BTN_STYLUS:
mBtnStylus = rawEvent->value;
break;
case BTN_STYLUS2:
mBtnStylus2 = rawEvent->value;
break;
case BTN_TOOL_FINGER:
mBtnToolFinger = rawEvent->value;
break;
case BTN_TOOL_PEN:
mBtnToolPen = rawEvent->value;
break;
case BTN_TOOL_RUBBER:
mBtnToolRubber = rawEvent->value;
break;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
case BTN_TOOL_BRUSH:
mBtnToolBrush = rawEvent->value;
break;
case BTN_TOOL_PENCIL:
mBtnToolPencil = rawEvent->value;
break;
case BTN_TOOL_AIRBRUSH:
mBtnToolAirbrush = rawEvent->value;
break;
case BTN_TOOL_MOUSE:
mBtnToolMouse = rawEvent->value;
break;
case BTN_TOOL_LENS:
mBtnToolLens = rawEvent->value;
break;
Support BTN_TOOL_*TAP as synonyms for BTN_TOOL_FINGER. Bug: 5205301 Some drivers report one finger as BTN_TOOL_FINGER, two as BTN_TOOL_DOUBLETAP, three as BTN_TOOL_TRIPLETAP and four as BTN_TOOL_QUADTAP. Since we care about the tool type, we need to handle _DOUBLE/_TRIPLE/_QUAD tap in the same way we handle _FINGER. Change-Id: I8eb83d2a2bada9ac32d07619c7eea84e924316b8
2011-08-23 17:31:25 -07:00
case BTN_TOOL_DOUBLETAP:
mBtnToolDoubleTap = rawEvent->value;
break;
case BTN_TOOL_TRIPLETAP:
mBtnToolTripleTap = rawEvent->value;
break;
case BTN_TOOL_QUADTAP:
mBtnToolQuadTap = rawEvent->value;
break;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
}
}
}
uint32_t TouchButtonAccumulator::getButtonState() const {
uint32_t result = 0;
if (mBtnStylus) {
result |= AMOTION_EVENT_BUTTON_SECONDARY;
}
if (mBtnStylus2) {
result |= AMOTION_EVENT_BUTTON_TERTIARY;
}
return result;
}
int32_t TouchButtonAccumulator::getToolType() const {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mBtnToolMouse || mBtnToolLens) {
return AMOTION_EVENT_TOOL_TYPE_MOUSE;
}
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
if (mBtnToolRubber) {
return AMOTION_EVENT_TOOL_TYPE_ERASER;
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mBtnToolPen || mBtnToolBrush || mBtnToolPencil || mBtnToolAirbrush) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
return AMOTION_EVENT_TOOL_TYPE_STYLUS;
}
Support BTN_TOOL_*TAP as synonyms for BTN_TOOL_FINGER. Bug: 5205301 Some drivers report one finger as BTN_TOOL_FINGER, two as BTN_TOOL_DOUBLETAP, three as BTN_TOOL_TRIPLETAP and four as BTN_TOOL_QUADTAP. Since we care about the tool type, we need to handle _DOUBLE/_TRIPLE/_QUAD tap in the same way we handle _FINGER. Change-Id: I8eb83d2a2bada9ac32d07619c7eea84e924316b8
2011-08-23 17:31:25 -07:00
if (mBtnToolFinger || mBtnToolDoubleTap || mBtnToolTripleTap || mBtnToolQuadTap) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
return AMOTION_EVENT_TOOL_TYPE_FINGER;
}
return AMOTION_EVENT_TOOL_TYPE_UNKNOWN;
}
Use BTN_TOUCH or BTN_TOOL_* to determine if touch active. Bug: 5064702 Should not include stylus buttons in the condition. Change-Id: If4d78a875b77da8bd59672d6fdbf5353004d0023
2011-08-10 14:55:59 -07:00
bool TouchButtonAccumulator::isToolActive() const {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
return mBtnTouch || mBtnToolFinger || mBtnToolPen || mBtnToolRubber
|| mBtnToolBrush || mBtnToolPencil || mBtnToolAirbrush
Support BTN_TOOL_*TAP as synonyms for BTN_TOOL_FINGER. Bug: 5205301 Some drivers report one finger as BTN_TOOL_FINGER, two as BTN_TOOL_DOUBLETAP, three as BTN_TOOL_TRIPLETAP and four as BTN_TOOL_QUADTAP. Since we care about the tool type, we need to handle _DOUBLE/_TRIPLE/_QUAD tap in the same way we handle _FINGER. Change-Id: I8eb83d2a2bada9ac32d07619c7eea84e924316b8
2011-08-23 17:31:25 -07:00
|| mBtnToolMouse || mBtnToolLens
|| mBtnToolDoubleTap || mBtnToolTripleTap || mBtnToolQuadTap;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
}
bool TouchButtonAccumulator::isHovering() const {
return mHaveBtnTouch && !mBtnTouch;
}
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
bool TouchButtonAccumulator::hasStylus() const {
return mHaveStylus;
}
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// --- RawPointerAxes ---
RawPointerAxes::RawPointerAxes() {
clear();
}
void RawPointerAxes::clear() {
x.clear();
y.clear();
pressure.clear();
touchMajor.clear();
touchMinor.clear();
toolMajor.clear();
toolMinor.clear();
orientation.clear();
distance.clear();
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
tiltX.clear();
tiltY.clear();
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
trackingId.clear();
slot.clear();
}
// --- RawPointerData ---
RawPointerData::RawPointerData() {
clear();
}
void RawPointerData::clear() {
pointerCount = 0;
clearIdBits();
}
void RawPointerData::copyFrom(const RawPointerData& other) {
pointerCount = other.pointerCount;
hoveringIdBits = other.hoveringIdBits;
touchingIdBits = other.touchingIdBits;
for (uint32_t i = 0; i < pointerCount; i++) {
pointers[i] = other.pointers[i];
int id = pointers[i].id;
idToIndex[id] = other.idToIndex[id];
}
}
void RawPointerData::getCentroidOfTouchingPointers(float* outX, float* outY) const {
float x = 0, y = 0;
uint32_t count = touchingIdBits.count();
if (count) {
for (BitSet32 idBits(touchingIdBits); !idBits.isEmpty(); ) {
uint32_t id = idBits.clearFirstMarkedBit();
const Pointer& pointer = pointerForId(id);
x += pointer.x;
y += pointer.y;
}
x /= count;
y /= count;
}
*outX = x;
*outY = y;
}
// --- CookedPointerData ---
CookedPointerData::CookedPointerData() {
clear();
}
void CookedPointerData::clear() {
pointerCount = 0;
hoveringIdBits.clear();
touchingIdBits.clear();
}
void CookedPointerData::copyFrom(const CookedPointerData& other) {
pointerCount = other.pointerCount;
hoveringIdBits = other.hoveringIdBits;
touchingIdBits = other.touchingIdBits;
for (uint32_t i = 0; i < pointerCount; i++) {
pointerProperties[i].copyFrom(other.pointerProperties[i]);
pointerCoords[i].copyFrom(other.pointerCoords[i]);
int id = pointerProperties[i].id;
idToIndex[id] = other.idToIndex[id];
}
}
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
// --- SingleTouchMotionAccumulator ---
SingleTouchMotionAccumulator::SingleTouchMotionAccumulator() {
clearAbsoluteAxes();
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void SingleTouchMotionAccumulator::reset(InputDevice* device) {
mAbsX = device->getAbsoluteAxisValue(ABS_X);
mAbsY = device->getAbsoluteAxisValue(ABS_Y);
mAbsPressure = device->getAbsoluteAxisValue(ABS_PRESSURE);
mAbsToolWidth = device->getAbsoluteAxisValue(ABS_TOOL_WIDTH);
mAbsDistance = device->getAbsoluteAxisValue(ABS_DISTANCE);
mAbsTiltX = device->getAbsoluteAxisValue(ABS_TILT_X);
mAbsTiltY = device->getAbsoluteAxisValue(ABS_TILT_Y);
}
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
void SingleTouchMotionAccumulator::clearAbsoluteAxes() {
mAbsX = 0;
mAbsY = 0;
mAbsPressure = 0;
mAbsToolWidth = 0;
mAbsDistance = 0;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mAbsTiltX = 0;
mAbsTiltY = 0;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
}
void SingleTouchMotionAccumulator::process(const RawEvent* rawEvent) {
if (rawEvent->type == EV_ABS) {
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
switch (rawEvent->code) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
case ABS_X:
mAbsX = rawEvent->value;
break;
case ABS_Y:
mAbsY = rawEvent->value;
break;
case ABS_PRESSURE:
mAbsPressure = rawEvent->value;
break;
case ABS_TOOL_WIDTH:
mAbsToolWidth = rawEvent->value;
break;
case ABS_DISTANCE:
mAbsDistance = rawEvent->value;
break;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
case ABS_TILT_X:
mAbsTiltX = rawEvent->value;
break;
case ABS_TILT_Y:
mAbsTiltY = rawEvent->value;
break;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
}
}
}
// --- MultiTouchMotionAccumulator ---
MultiTouchMotionAccumulator::MultiTouchMotionAccumulator() :
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
mCurrentSlot(-1), mSlots(NULL), mSlotCount(0), mUsingSlotsProtocol(false),
mHaveStylus(false) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
}
MultiTouchMotionAccumulator::~MultiTouchMotionAccumulator() {
delete[] mSlots;
}
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
void MultiTouchMotionAccumulator::configure(InputDevice* device,
size_t slotCount, bool usingSlotsProtocol) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
mSlotCount = slotCount;
mUsingSlotsProtocol = usingSlotsProtocol;
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
mHaveStylus = device->hasAbsoluteAxis(ABS_MT_TOOL_TYPE);
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
delete[] mSlots;
mSlots = new Slot[slotCount];
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void MultiTouchMotionAccumulator::reset(InputDevice* device) {
// Unfortunately there is no way to read the initial contents of the slots.
// So when we reset the accumulator, we must assume they are all zeroes.
if (mUsingSlotsProtocol) {
// Query the driver for the current slot index and use it as the initial slot
// before we start reading events from the device. It is possible that the
// current slot index will not be the same as it was when the first event was
// written into the evdev buffer, which means the input mapper could start
// out of sync with the initial state of the events in the evdev buffer.
// In the extremely unlikely case that this happens, the data from
// two slots will be confused until the next ABS_MT_SLOT event is received.
// This can cause the touch point to "jump", but at least there will be
// no stuck touches.
int32_t initialSlot;
status_t status = device->getEventHub()->getAbsoluteAxisValue(device->getId(),
ABS_MT_SLOT, &initialSlot);
if (status) {
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Could not retrieve current multitouch slot index. status=%d", status);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
initialSlot = -1;
}
clearSlots(initialSlot);
} else {
clearSlots(-1);
}
}
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
void MultiTouchMotionAccumulator::clearSlots(int32_t initialSlot) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mSlots) {
for (size_t i = 0; i < mSlotCount; i++) {
mSlots[i].clear();
}
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
}
mCurrentSlot = initialSlot;
}
void MultiTouchMotionAccumulator::process(const RawEvent* rawEvent) {
if (rawEvent->type == EV_ABS) {
bool newSlot = false;
if (mUsingSlotsProtocol) {
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
if (rawEvent->code == ABS_MT_SLOT) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
mCurrentSlot = rawEvent->value;
newSlot = true;
}
} else if (mCurrentSlot < 0) {
mCurrentSlot = 0;
}
if (mCurrentSlot < 0 || size_t(mCurrentSlot) >= mSlotCount) {
#if DEBUG_POINTERS
if (newSlot) {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW("MultiTouch device emitted invalid slot index %d but it "
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
"should be between 0 and %d; ignoring this slot.",
mCurrentSlot, mSlotCount - 1);
}
#endif
} else {
Slot* slot = &mSlots[mCurrentSlot];
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
switch (rawEvent->code) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
case ABS_MT_POSITION_X:
slot->mInUse = true;
slot->mAbsMTPositionX = rawEvent->value;
break;
case ABS_MT_POSITION_Y:
slot->mInUse = true;
slot->mAbsMTPositionY = rawEvent->value;
break;
case ABS_MT_TOUCH_MAJOR:
slot->mInUse = true;
slot->mAbsMTTouchMajor = rawEvent->value;
break;
case ABS_MT_TOUCH_MINOR:
slot->mInUse = true;
slot->mAbsMTTouchMinor = rawEvent->value;
slot->mHaveAbsMTTouchMinor = true;
break;
case ABS_MT_WIDTH_MAJOR:
slot->mInUse = true;
slot->mAbsMTWidthMajor = rawEvent->value;
break;
case ABS_MT_WIDTH_MINOR:
slot->mInUse = true;
slot->mAbsMTWidthMinor = rawEvent->value;
slot->mHaveAbsMTWidthMinor = true;
break;
case ABS_MT_ORIENTATION:
slot->mInUse = true;
slot->mAbsMTOrientation = rawEvent->value;
break;
case ABS_MT_TRACKING_ID:
if (mUsingSlotsProtocol && rawEvent->value < 0) {
Exorcise notification panel ghosts. Bug: 5105599 Don't clear slot data when slot is no longer in use. The kernel will not resend absolute slot values if they are unchanged, even if the slot was previously marked unused. Because the protocol is stateful, this could cause problems if evdev drops events (marked by SYN_DROPPED) since we cannot query the initial values of the slots, but there's nothing we can do. Change-Id: Ie13e68097fb8ed2542c2f60338b499082cf9e77f
2011-08-11 15:49:09 -07:00
// The slot is no longer in use but it retains its previous contents,
// which may be reused for subsequent touches.
slot->mInUse = false;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
} else {
slot->mInUse = true;
slot->mAbsMTTrackingId = rawEvent->value;
}
break;
case ABS_MT_PRESSURE:
slot->mInUse = true;
slot->mAbsMTPressure = rawEvent->value;
break;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
case ABS_MT_DISTANCE:
slot->mInUse = true;
slot->mAbsMTDistance = rawEvent->value;
break;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
case ABS_MT_TOOL_TYPE:
slot->mInUse = true;
slot->mAbsMTToolType = rawEvent->value;
slot->mHaveAbsMTToolType = true;
break;
}
}
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
} else if (rawEvent->type == EV_SYN && rawEvent->code == SYN_MT_REPORT) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
// MultiTouch Sync: The driver has returned all data for *one* of the pointers.
mCurrentSlot += 1;
}
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void MultiTouchMotionAccumulator::finishSync() {
if (!mUsingSlotsProtocol) {
clearSlots(-1);
}
}
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
bool MultiTouchMotionAccumulator::hasStylus() const {
return mHaveStylus;
}
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
// --- MultiTouchMotionAccumulator::Slot ---
MultiTouchMotionAccumulator::Slot::Slot() {
clear();
}
void MultiTouchMotionAccumulator::Slot::clear() {
mInUse = false;
mHaveAbsMTTouchMinor = false;
mHaveAbsMTWidthMinor = false;
mHaveAbsMTToolType = false;
mAbsMTPositionX = 0;
mAbsMTPositionY = 0;
mAbsMTTouchMajor = 0;
mAbsMTTouchMinor = 0;
mAbsMTWidthMajor = 0;
mAbsMTWidthMinor = 0;
mAbsMTOrientation = 0;
mAbsMTTrackingId = -1;
mAbsMTPressure = 0;
mAbsMTDistance = 0;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mAbsMTToolType = 0;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
}
int32_t MultiTouchMotionAccumulator::Slot::getToolType() const {
if (mHaveAbsMTToolType) {
switch (mAbsMTToolType) {
case MT_TOOL_FINGER:
return AMOTION_EVENT_TOOL_TYPE_FINGER;
case MT_TOOL_PEN:
return AMOTION_EVENT_TOOL_TYPE_STYLUS;
}
}
return AMOTION_EVENT_TOOL_TYPE_UNKNOWN;
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// --- InputMapper ---
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
InputMapper::InputMapper(InputDevice* device) :
mDevice(device), mContext(device->getContext()) {
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
InputMapper::~InputMapper() {
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
void InputMapper::populateDeviceInfo(InputDeviceInfo* info) {
info->addSource(getSources());
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
void InputMapper::dump(String8& dump) {
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void InputMapper::configure(nsecs_t when,
const InputReaderConfiguration* config, uint32_t changes) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void InputMapper::reset(nsecs_t when) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor how timeouts are calculated. Added a timeout mechanism to EventHub and InputReader so that InputMappers can request timeouts to perform delayed processing of input when needed. Change-Id: Iec2045baaf4e67690b15eef3c09a58d5cac76897
2011-03-17 01:34:19 -07:00
void InputMapper::timeoutExpired(nsecs_t when) {
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t InputMapper::getKeyCodeState(uint32_t sourceMask, int32_t keyCode) {
return AKEY_STATE_UNKNOWN;
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t InputMapper::getScanCodeState(uint32_t sourceMask, int32_t scanCode) {
return AKEY_STATE_UNKNOWN;
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t InputMapper::getSwitchState(uint32_t sourceMask, int32_t switchCode) {
return AKEY_STATE_UNKNOWN;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
bool InputMapper::markSupportedKeyCodes(uint32_t sourceMask, size_t numCodes,
const int32_t* keyCodes, uint8_t* outFlags) {
return false;
}
Native input event dispatching. Target identification is now fully native. Fixed a couple of minor issues related to input injection. Native input enabled by default, can be disabled by setting WindowManagerPolicy.ENABLE_NATIVE_INPUT_DISPATCH to false. Change-Id: I7edf66ed3e987cc9306ad4743ac57a116af452ff
2010-06-22 01:27:15 -07:00
Add support for input devices that have vibrators. Added a getVibrator() method to InputDevice which returns a Vibrator associated with that input device. Its uses the same API as the system vibrator which makes it easy for applications to be modified to use one or the other. Bug: 6334179 Change-Id: Ifc7f13dbcb778670f3f1c07ccc562334e6109d2e
2012-04-13 04:09:27 -07:00
void InputMapper::vibrate(const nsecs_t* pattern, size_t patternSize, ssize_t repeat,
int32_t token) {
}
void InputMapper::cancelVibrate(int32_t token) {
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t InputMapper::getMetaState() {
return 0;
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Fade out the mouse pointer after inactivity or other events. Fades out the mouse pointer: - after 15 seconds of inactivity normally - after 3 seconds of inactivity in lights out mode - after a non-modifier key down - after a touch down Extended the native Looper to support enqueuing time delayed messages. This is used by the PointerController to control pointer fade timing. Change-Id: I87792fea7dbe2d9376c78cf354fe3189a484d9da
2011-03-02 14:41:58 -08:00
void InputMapper::fadePointer() {
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
status_t InputMapper::getAbsoluteAxisInfo(int32_t axis, RawAbsoluteAxisInfo* axisInfo) {
return getEventHub()->getAbsoluteAxisInfo(getDeviceId(), axis, axisInfo);
}
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
void InputMapper::bumpGeneration() {
mDevice->bumpGeneration();
}
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
void InputMapper::dumpRawAbsoluteAxisInfo(String8& dump,
const RawAbsoluteAxisInfo& axis, const char* name) {
if (axis.valid) {
Take advantage of updated linux/input.h kernel headers. Change-Id: I72d2ef82de5c504d46b0cdb57aa43bbd0d769174
2011-06-12 18:14:50 -07:00
dump.appendFormat(INDENT4 "%s: min=%d, max=%d, flat=%d, fuzz=%d, resolution=%d\n",
name, axis.minValue, axis.maxValue, axis.flat, axis.fuzz, axis.resolution);
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
} else {
dump.appendFormat(INDENT4 "%s: unknown range\n", name);
}
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// --- SwitchInputMapper ---
SwitchInputMapper::SwitchInputMapper(InputDevice* device) :
InputMapper(device) {
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
SwitchInputMapper::~SwitchInputMapper() {
}
uint32_t SwitchInputMapper::getSources() {
Fix getSwitchState. InputReader::getSwitchState always returns AKEY_STATE_UNKNOWN because SwitchInputMapper::getSources() returns 0 which cannot match any source mask including AINPUT_SOURCE_ANY. As a result initial lid switch detection is broken. This change adds a new source constant AINPUT_SOURCE_SWITCH that indicates that the source has switches. Change-Id: I5321ecf0ce84f1c2b4535f6c163d3f4dcf9b7a9b
2011-01-19 18:41:38 -08:00
return AINPUT_SOURCE_SWITCH;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
void SwitchInputMapper::process(const RawEvent* rawEvent) {
switch (rawEvent->type) {
case EV_SW:
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
processSwitch(rawEvent->when, rawEvent->code, rawEvent->value);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
break;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
void SwitchInputMapper::processSwitch(nsecs_t when, int32_t switchCode, int32_t switchValue) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
NotifySwitchArgs args(when, 0, switchCode, switchValue);
getListener()->notifySwitch(&args);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t SwitchInputMapper::getSwitchState(uint32_t sourceMask, int32_t switchCode) {
return getEventHub()->getSwitchState(getDeviceId(), switchCode);
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Add support for input devices that have vibrators. Added a getVibrator() method to InputDevice which returns a Vibrator associated with that input device. Its uses the same API as the system vibrator which makes it easy for applications to be modified to use one or the other. Bug: 6334179 Change-Id: Ifc7f13dbcb778670f3f1c07ccc562334e6109d2e
2012-04-13 04:09:27 -07:00
// --- VibratorInputMapper ---
VibratorInputMapper::VibratorInputMapper(InputDevice* device) :
InputMapper(device), mVibrating(false) {
}
VibratorInputMapper::~VibratorInputMapper() {
}
uint32_t VibratorInputMapper::getSources() {
return 0;
}
void VibratorInputMapper::populateDeviceInfo(InputDeviceInfo* info) {
InputMapper::populateDeviceInfo(info);
info->setVibrator(true);
}
void VibratorInputMapper::process(const RawEvent* rawEvent) {
// TODO: Handle FF_STATUS, although it does not seem to be widely supported.
}
void VibratorInputMapper::vibrate(const nsecs_t* pattern, size_t patternSize, ssize_t repeat,
int32_t token) {
#if DEBUG_VIBRATOR
String8 patternStr;
for (size_t i = 0; i < patternSize; i++) {
if (i != 0) {
patternStr.append(", ");
}
patternStr.appendFormat("%lld", pattern[i]);
}
ALOGD("vibrate: deviceId=%d, pattern=[%s], repeat=%ld, token=%d",
getDeviceId(), patternStr.string(), repeat, token);
#endif
mVibrating = true;
memcpy(mPattern, pattern, patternSize * sizeof(nsecs_t));
mPatternSize = patternSize;
mRepeat = repeat;
mToken = token;
mIndex = -1;
nextStep();
}
void VibratorInputMapper::cancelVibrate(int32_t token) {
#if DEBUG_VIBRATOR
ALOGD("cancelVibrate: deviceId=%d, token=%d", getDeviceId(), token);
#endif
if (mVibrating && mToken == token) {
stopVibrating();
}
}
void VibratorInputMapper::timeoutExpired(nsecs_t when) {
if (mVibrating) {
if (when >= mNextStepTime) {
nextStep();
} else {
getContext()->requestTimeoutAtTime(mNextStepTime);
}
}
}
void VibratorInputMapper::nextStep() {
mIndex += 1;
if (size_t(mIndex) >= mPatternSize) {
if (mRepeat < 0) {
// We are done.
stopVibrating();
return;
}
mIndex = mRepeat;
}
bool vibratorOn = mIndex & 1;
nsecs_t duration = mPattern[mIndex];
if (vibratorOn) {
#if DEBUG_VIBRATOR
ALOGD("nextStep: sending vibrate deviceId=%d, duration=%lld",
getDeviceId(), duration);
#endif
getEventHub()->vibrate(getDeviceId(), duration);
} else {
#if DEBUG_VIBRATOR
ALOGD("nextStep: sending cancel vibrate deviceId=%d", getDeviceId());
#endif
getEventHub()->cancelVibrate(getDeviceId());
}
nsecs_t now = systemTime(SYSTEM_TIME_MONOTONIC);
mNextStepTime = now + duration;
getContext()->requestTimeoutAtTime(mNextStepTime);
#if DEBUG_VIBRATOR
ALOGD("nextStep: scheduled timeout in %0.3fms", duration * 0.000001f);
#endif
}
void VibratorInputMapper::stopVibrating() {
mVibrating = false;
#if DEBUG_VIBRATOR
ALOGD("stopVibrating: sending cancel vibrate deviceId=%d", getDeviceId());
#endif
getEventHub()->cancelVibrate(getDeviceId());
}
void VibratorInputMapper::dump(String8& dump) {
dump.append(INDENT2 "Vibrator Input Mapper:\n");
dump.appendFormat(INDENT3 "Vibrating: %s\n", toString(mVibrating));
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// --- KeyboardInputMapper ---
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
KeyboardInputMapper::KeyboardInputMapper(InputDevice* device,
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
uint32_t source, int32_t keyboardType) :
InputMapper(device), mSource(source),
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
mKeyboardType(keyboardType) {
}
KeyboardInputMapper::~KeyboardInputMapper() {
}
uint32_t KeyboardInputMapper::getSources() {
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
return mSource;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
void KeyboardInputMapper::populateDeviceInfo(InputDeviceInfo* info) {
InputMapper::populateDeviceInfo(info);
info->setKeyboardType(mKeyboardType);
Request key maps from input manager service. Instead of each application loading the KeyCharacterMap from the file system, get them from the input manager service as part of the InputDevice object. Refactored InputManager to be a proper singleton instead of having a bunch of static methods. InputManager now maintains a cache of all InputDevice objects that it has loaded. Currently we never invalidate the cache which can cause InputDevice to return stale motion ranges if the device is reconfigured. This will be fixed in a future change. Added a fake InputDevice with ID -1 to represent the virtual keyboard. Change-Id: If7a695839ad0972317a5aab89e9d1e42ace28eb7
2012-04-10 14:30:49 -07:00
info->setKeyCharacterMap(getEventHub()->getKeyCharacterMap(getDeviceId()));
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
void KeyboardInputMapper::dump(String8& dump) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.append(INDENT2 "Keyboard Input Mapper:\n");
dumpParameters(dump);
dump.appendFormat(INDENT3 "KeyboardType: %d\n", mKeyboardType);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dump.appendFormat(INDENT3 "Orientation: %d\n", mOrientation);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.appendFormat(INDENT3 "KeyDowns: %d keys currently down\n", mKeyDowns.size());
dump.appendFormat(INDENT3 "MetaState: 0x%0x\n", mMetaState);
dump.appendFormat(INDENT3 "DownTime: %lld\n", mDownTime);
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
}
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void KeyboardInputMapper::configure(nsecs_t when,
const InputReaderConfiguration* config, uint32_t changes) {
InputMapper::configure(when, config, changes);
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (!changes) { // first time only
// Configure basic parameters.
configureParameters();
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
}
Add support for fallback keycodes. This change enables the framework to synthesize key events to implement default behavior when an application does not handle a key. For example, this change enables numeric keypad keys to perform their associated special function when numlock is off. The application is informed that it is processing a fallback keypress so it can choose to ignore it. Added a new keycode for switching applications. Added ALT key deadkeys. New default key mappings: - ESC -> BACK - Meta+ESC -> HOME - Alt+ESC -> MENU - Meta+Space -> SEARCH - Meta+Tab -> APP_SWITCH Fixed some comments. Fixed some tests. Change-Id: Id7f3b6645f3a350275e624547822f72652f3defe
2010-12-06 17:13:33 -08:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (!changes || (changes & InputReaderConfiguration::CHANGE_DISPLAY_INFO)) {
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
if (mParameters.orientationAware && mParameters.hasAssociatedDisplay) {
DisplayViewport v;
if (config->getDisplayInfo(false /*external*/, &v)) {
mOrientation = v.orientation;
} else {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mOrientation = DISPLAY_ORIENTATION_0;
}
} else {
mOrientation = DISPLAY_ORIENTATION_0;
}
Add support for fallback keycodes. This change enables the framework to synthesize key events to implement default behavior when an application does not handle a key. For example, this change enables numeric keypad keys to perform their associated special function when numlock is off. The application is informed that it is processing a fallback keypress so it can choose to ignore it. Added a new keycode for switching applications. Added ALT key deadkeys. New default key mappings: - ESC -> BACK - Meta+ESC -> HOME - Alt+ESC -> MENU - Meta+Space -> SEARCH - Meta+Tab -> APP_SWITCH Fixed some comments. Fixed some tests. Change-Id: Id7f3b6645f3a350275e624547822f72652f3defe
2010-12-06 17:13:33 -08:00
}
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
}
void KeyboardInputMapper::configureParameters() {
mParameters.orientationAware = false;
getDevice()->getConfiguration().tryGetProperty(String8("keyboard.orientationAware"),
mParameters.orientationAware);
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
mParameters.hasAssociatedDisplay = false;
Report the external display size to the input reader. The input reader needs this information so that it knows how to interpolate touches on an external touch screen. Changed Display so that it asks the WindowManager what the real display size is (as opposed to the raw display size). This means it now takes into the forced display size set by adb shell am display-size. Replaced all calls to getRealWidth() / getRealHeight() / getRealMetrics() in the WindowManager and replaced them with direct usages of the mCurDisplayWidth / mCurDisplayHeight so that the WM doesn't end up making a reentrant Binder call into itself. Fixed the table status bar HeightReceiver so that it updates the height on all configuration changes since it is possible that the display size changed independently of an external HDMI display being plugged / unplugged. Improved the Display class documentation to make the distinctions betweeen the various sizes clearer. Change-Id: I3f75de559d3ebffed532ab46c4ae52c5e7f1da2b
2011-07-25 12:58:12 -07:00
if (mParameters.orientationAware) {
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
mParameters.hasAssociatedDisplay = true;
Report the external display size to the input reader. The input reader needs this information so that it knows how to interpolate touches on an external touch screen. Changed Display so that it asks the WindowManager what the real display size is (as opposed to the raw display size). This means it now takes into the forced display size set by adb shell am display-size. Replaced all calls to getRealWidth() / getRealHeight() / getRealMetrics() in the WindowManager and replaced them with direct usages of the mCurDisplayWidth / mCurDisplayHeight so that the WM doesn't end up making a reentrant Binder call into itself. Fixed the table status bar HeightReceiver so that it updates the height on all configuration changes since it is possible that the display size changed independently of an external HDMI display being plugged / unplugged. Improved the Display class documentation to make the distinctions betweeen the various sizes clearer. Change-Id: I3f75de559d3ebffed532ab46c4ae52c5e7f1da2b
2011-07-25 12:58:12 -07:00
}
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
}
void KeyboardInputMapper::dumpParameters(String8& dump) {
dump.append(INDENT3 "Parameters:\n");
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
dump.appendFormat(INDENT4 "HasAssociatedDisplay: %s\n",
toString(mParameters.hasAssociatedDisplay));
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
dump.appendFormat(INDENT4 "OrientationAware: %s\n",
toString(mParameters.orientationAware));
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void KeyboardInputMapper::reset(nsecs_t when) {
mMetaState = AMETA_NONE;
mDownTime = 0;
mKeyDowns.clear();
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
mCurrentHidUsage = 0;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
resetLedState();
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
InputMapper::reset(when);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
void KeyboardInputMapper::process(const RawEvent* rawEvent) {
switch (rawEvent->type) {
case EV_KEY: {
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
int32_t scanCode = rawEvent->code;
int32_t usageCode = mCurrentHidUsage;
mCurrentHidUsage = 0;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
if (isKeyboardOrGamepadKey(scanCode)) {
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
int32_t keyCode;
uint32_t flags;
if (getEventHub()->mapKey(getDeviceId(), scanCode, usageCode, &keyCode, &flags)) {
keyCode = AKEYCODE_UNKNOWN;
flags = 0;
}
processKey(rawEvent->when, rawEvent->value != 0, keyCode, scanCode, flags);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
break;
}
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
case EV_MSC: {
if (rawEvent->code == MSC_SCAN) {
mCurrentHidUsage = rawEvent->value;
}
break;
}
case EV_SYN: {
if (rawEvent->code == SYN_REPORT) {
mCurrentHidUsage = 0;
}
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
bool KeyboardInputMapper::isKeyboardOrGamepadKey(int32_t scanCode) {
return scanCode < BTN_MOUSE
|| scanCode >= KEY_OK
Recognize 3D joysticks. (DO NOT MERGE) Change-Id: Idf7799ea90fa07677850d40ee1aec39220ded372
2011-03-03 03:39:29 -08:00
|| (scanCode >= BTN_MISC && scanCode < BTN_MOUSE)
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
|| (scanCode >= BTN_JOYSTICK && scanCode < BTN_DIGI);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
void KeyboardInputMapper::processKey(nsecs_t when, bool down, int32_t keyCode,
int32_t scanCode, uint32_t policyFlags) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (down) {
// Rotate key codes according to orientation if needed.
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
if (mParameters.orientationAware && mParameters.hasAssociatedDisplay) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
keyCode = rotateKeyCode(keyCode, mOrientation);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// Add key down.
ssize_t keyDownIndex = findKeyDown(scanCode);
if (keyDownIndex >= 0) {
// key repeat, be sure to use same keycode as before in case of rotation
keyCode = mKeyDowns.itemAt(keyDownIndex).keyCode;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
} else {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// key down
if ((policyFlags & POLICY_FLAG_VIRTUAL)
&& mContext->shouldDropVirtualKey(when,
getDevice(), keyCode, scanCode)) {
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
return;
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mKeyDowns.push();
KeyDown& keyDown = mKeyDowns.editTop();
keyDown.keyCode = keyCode;
keyDown.scanCode = scanCode;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mDownTime = when;
} else {
// Remove key down.
ssize_t keyDownIndex = findKeyDown(scanCode);
if (keyDownIndex >= 0) {
// key up, be sure to use same keycode as before in case of rotation
keyCode = mKeyDowns.itemAt(keyDownIndex).keyCode;
mKeyDowns.removeAt(size_t(keyDownIndex));
} else {
// key was not actually down
Rename (IF_)LOGI(_IF) to (IF_)ALOGI(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156801 Bug: 5449033 Change-Id: Ib08fe86d23db91ee153e9f91a99a35c42b9208ea
2012-01-04 20:05:49 +00:00
ALOGI("Dropping key up from device %s because the key was not down. "
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
"keyCode=%d, scanCode=%d",
getDeviceName().string(), keyCode, scanCode);
return;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
bool metaStateChanged = false;
int32_t oldMetaState = mMetaState;
int32_t newMetaState = updateMetaState(keyCode, down, oldMetaState);
if (oldMetaState != newMetaState) {
mMetaState = newMetaState;
metaStateChanged = true;
updateLedState(false);
}
nsecs_t downTime = mDownTime;
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
Wake screen from external HID peripherals. Added some plumbing to enable the policy to intercept motion events when the screen is off to handle wakeup if needed. Added a basic concept of an external device to limit the scope of the wakeup policy to external devices only. The wakeup policy for internal devices should be based on explicit rules such as policy flags in key layout files. Moved isTouchEvent to native. Ensure the dispatcher sends the right event type to userActivity for non-touch pointer events like HOVER_MOVE and SCROLL. Bug: 3193114 Change-Id: I15dbd48a16810dfaf226ff7ad117d46908ca4f86
2011-03-02 19:23:13 -08:00
// Key down on external an keyboard should wake the device.
// We don't do this for internal keyboards to prevent them from waking up in your pocket.
// For internal keyboards, the key layout file should specify the policy flags for
// each wake key individually.
// TODO: Use the input device configuration to control this behavior more finely.
if (down && getDevice()->isExternal()
&& !(policyFlags & (POLICY_FLAG_WAKE | POLICY_FLAG_WAKE_DROPPED))) {
policyFlags |= POLICY_FLAG_WAKE_DROPPED;
}
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
if (metaStateChanged) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
getContext()->updateGlobalMetaState();
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Fade out the mouse pointer after inactivity or other events. Fades out the mouse pointer: - after 15 seconds of inactivity normally - after 3 seconds of inactivity in lights out mode - after a non-modifier key down - after a touch down Extended the native Looper to support enqueuing time delayed messages. This is used by the PointerController to control pointer fade timing. Change-Id: I87792fea7dbe2d9376c78cf354fe3189a484d9da
2011-03-02 14:41:58 -08:00
if (down && !isMetaKey(keyCode)) {
getContext()->fadePointer();
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
NotifyKeyArgs args(when, getDeviceId(), mSource, policyFlags,
Added more robust tracking and cancelation of events. This change fixes several issues where events would be dropped in the input dispatch pipeline in such a way that the dispatcher could not accurately track the state of the input device. Given more robust tracking, we can now also provide robust cancelation of input events in cases where an application might otherwise become out of sync with the event stream due to ANR, app switch, policy decisions, or forced focus transitions. Pruned some of the input dispatcher log output. Moved the responsibility for calling intercept*BeforeQueueing into the input dispatcher instead of the input reader and added support for early interception of injected events for events coming from trusted sources. This enables behaviors like injection of media keys while the screen is off, haptic feedback of injected virtual keys, so injected events become more "first class" in a way. Change-Id: Iec6ff1dd21e5f3c7feb80ea4feb5382bd090dbd9
2010-10-08 22:31:17 -07:00
down ? AKEY_EVENT_ACTION_DOWN : AKEY_EVENT_ACTION_UP,
AKEY_EVENT_FLAG_FROM_SYSTEM, keyCode, scanCode, newMetaState, downTime);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
getListener()->notifyKey(&args);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Add support for new input sources. Added several new coordinate values to MotionEvents to capture touch major/minor area, tool major/minor area and orientation. Renamed NDK input constants per convention. Added InputDevice class in Java which will eventually provide useful information about available input devices. Added APIs for manufacturing new MotionEvent objects with multiple pointers and all necessary coordinate data. Fixed a bug in the input dispatcher where it could get stuck with a pointer down forever. Fixed a bug in the WindowManager where the input window list could end up containing stale removed windows. Fixed a bug in the WindowManager where the input channel was being removed only after the final animation transition had taken place which caused spurious WINDOW DIED log messages to be printed. Change-Id: Ie55084da319b20aad29b28a0499b8dd98bb5da68
2010-07-14 18:48:53 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
ssize_t KeyboardInputMapper::findKeyDown(int32_t scanCode) {
size_t n = mKeyDowns.size();
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
for (size_t i = 0; i < n; i++) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (mKeyDowns[i].scanCode == scanCode) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
return i;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
return -1;
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t KeyboardInputMapper::getKeyCodeState(uint32_t sourceMask, int32_t keyCode) {
return getEventHub()->getKeyCodeState(getDeviceId(), keyCode);
}
Add support for new input sources. Added several new coordinate values to MotionEvents to capture touch major/minor area, tool major/minor area and orientation. Renamed NDK input constants per convention. Added InputDevice class in Java which will eventually provide useful information about available input devices. Added APIs for manufacturing new MotionEvent objects with multiple pointers and all necessary coordinate data. Fixed a bug in the input dispatcher where it could get stuck with a pointer down forever. Fixed a bug in the WindowManager where the input window list could end up containing stale removed windows. Fixed a bug in the WindowManager where the input channel was being removed only after the final animation transition had taken place which caused spurious WINDOW DIED log messages to be printed. Change-Id: Ie55084da319b20aad29b28a0499b8dd98bb5da68
2010-07-14 18:48:53 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t KeyboardInputMapper::getScanCodeState(uint32_t sourceMask, int32_t scanCode) {
return getEventHub()->getScanCodeState(getDeviceId(), scanCode);
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
bool KeyboardInputMapper::markSupportedKeyCodes(uint32_t sourceMask, size_t numCodes,
const int32_t* keyCodes, uint8_t* outFlags) {
return getEventHub()->markSupportedKeyCodes(getDeviceId(), numCodes, keyCodes, outFlags);
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t KeyboardInputMapper::getMetaState() {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
return mMetaState;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Fix individual pointer id up/down reporting. Fix a minor threading bug in InputManager dump. Change-Id: Ic2eecf7df5a8dc9f40561fcb03ebe58a2c073778
2010-07-16 15:01:56 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void KeyboardInputMapper::resetLedState() {
initializeLedState(mCapsLockLedState, LED_CAPSL);
initializeLedState(mNumLockLedState, LED_NUML);
initializeLedState(mScrollLockLedState, LED_SCROLLL);
Add support for fallback keycodes. This change enables the framework to synthesize key events to implement default behavior when an application does not handle a key. For example, this change enables numeric keypad keys to perform their associated special function when numlock is off. The application is informed that it is processing a fallback keypress so it can choose to ignore it. Added a new keycode for switching applications. Added ALT key deadkeys. New default key mappings: - ESC -> BACK - Meta+ESC -> HOME - Alt+ESC -> MENU - Meta+Space -> SEARCH - Meta+Tab -> APP_SWITCH Fixed some comments. Fixed some tests. Change-Id: Id7f3b6645f3a350275e624547822f72652f3defe
2010-12-06 17:13:33 -08:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
updateLedState(true);
Add support for fallback keycodes. This change enables the framework to synthesize key events to implement default behavior when an application does not handle a key. For example, this change enables numeric keypad keys to perform their associated special function when numlock is off. The application is informed that it is processing a fallback keypress so it can choose to ignore it. Added a new keycode for switching applications. Added ALT key deadkeys. New default key mappings: - ESC -> BACK - Meta+ESC -> HOME - Alt+ESC -> MENU - Meta+Space -> SEARCH - Meta+Tab -> APP_SWITCH Fixed some comments. Fixed some tests. Change-Id: Id7f3b6645f3a350275e624547822f72652f3defe
2010-12-06 17:13:33 -08:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void KeyboardInputMapper::initializeLedState(LedState& ledState, int32_t led) {
Add support for fallback keycodes. This change enables the framework to synthesize key events to implement default behavior when an application does not handle a key. For example, this change enables numeric keypad keys to perform their associated special function when numlock is off. The application is informed that it is processing a fallback keypress so it can choose to ignore it. Added a new keycode for switching applications. Added ALT key deadkeys. New default key mappings: - ESC -> BACK - Meta+ESC -> HOME - Alt+ESC -> MENU - Meta+Space -> SEARCH - Meta+Tab -> APP_SWITCH Fixed some comments. Fixed some tests. Change-Id: Id7f3b6645f3a350275e624547822f72652f3defe
2010-12-06 17:13:33 -08:00
ledState.avail = getEventHub()->hasLed(getDeviceId(), led);
ledState.on = false;
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void KeyboardInputMapper::updateLedState(bool reset) {
updateLedStateForModifier(mCapsLockLedState, LED_CAPSL,
Rename the locked meta key constants for clarity. Also added some tests for LED setting. Change-Id: I3fd86322afd07ae8de52d1ccbc2fae2c6d586641
2010-10-29 22:19:53 -07:00
AMETA_CAPS_LOCK_ON, reset);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
updateLedStateForModifier(mNumLockLedState, LED_NUML,
Rename the locked meta key constants for clarity. Also added some tests for LED setting. Change-Id: I3fd86322afd07ae8de52d1ccbc2fae2c6d586641
2010-10-29 22:19:53 -07:00
AMETA_NUM_LOCK_ON, reset);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
updateLedStateForModifier(mScrollLockLedState, LED_SCROLLL,
Rename the locked meta key constants for clarity. Also added some tests for LED setting. Change-Id: I3fd86322afd07ae8de52d1ccbc2fae2c6d586641
2010-10-29 22:19:53 -07:00
AMETA_SCROLL_LOCK_ON, reset);
Add keycodes and meta-key modifiers to support external keyboards. Added new key maps for external keyboards. These maps are intended to be shared across devices by inheriting the "keyboards.mk" product makefile as part of the device's product definition. One of the trickier changes here was to unwind some code in MetaKeyKeyListener that assumed that only the low 8 bits of the meta key state were actually used. The new code abandons bitshifts in favor of simple conditionals that are probably easier to read anyways. The special meta key state constants used by MetaKeyKeyListener are now (@hide) defined in KeyEvent now so as to make it clearer that they share the same code space even if those codes are not valid for KeyEvents. The EventHub now takes care of detecting the appropriate key layout map and key character map when the device is added and sets system properties accordingly. This avoids having duplicate code in KeyCharacterMap to probe for the appropriate key character map although the current probing mechanism has been preserved for legacy reasons just in case. Added support for tracking caps lock, num lock and scroll lock and turning their corresponding LEDs on and off as needed. The key character map format will need to be updated to correctly support PC style external keyboard semantics related to modifier keys. That will come in a later change so caps lock doesn't actually do anything right now except turn the shiny LEDs on and off... Added a list of symbolic key names to KeyEvent and improved the toString() output for debug diagnosis. Having this list in a central place in the framework also allows us to remove it from Monkey so there is one less thing to maintain when we add new keycodes. Bug: 2912307 Change-Id: If8c25e8d50a7c29bbf5d663c94284f5f86de5da4
2010-09-12 17:55:08 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void KeyboardInputMapper::updateLedStateForModifier(LedState& ledState,
Add keycodes and meta-key modifiers to support external keyboards. Added new key maps for external keyboards. These maps are intended to be shared across devices by inheriting the "keyboards.mk" product makefile as part of the device's product definition. One of the trickier changes here was to unwind some code in MetaKeyKeyListener that assumed that only the low 8 bits of the meta key state were actually used. The new code abandons bitshifts in favor of simple conditionals that are probably easier to read anyways. The special meta key state constants used by MetaKeyKeyListener are now (@hide) defined in KeyEvent now so as to make it clearer that they share the same code space even if those codes are not valid for KeyEvents. The EventHub now takes care of detecting the appropriate key layout map and key character map when the device is added and sets system properties accordingly. This avoids having duplicate code in KeyCharacterMap to probe for the appropriate key character map although the current probing mechanism has been preserved for legacy reasons just in case. Added support for tracking caps lock, num lock and scroll lock and turning their corresponding LEDs on and off as needed. The key character map format will need to be updated to correctly support PC style external keyboard semantics related to modifier keys. That will come in a later change so caps lock doesn't actually do anything right now except turn the shiny LEDs on and off... Added a list of symbolic key names to KeyEvent and improved the toString() output for debug diagnosis. Having this list in a central place in the framework also allows us to remove it from Monkey so there is one less thing to maintain when we add new keycodes. Bug: 2912307 Change-Id: If8c25e8d50a7c29bbf5d663c94284f5f86de5da4
2010-09-12 17:55:08 -07:00
int32_t led, int32_t modifier, bool reset) {
if (ledState.avail) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
bool desiredState = (mMetaState & modifier) != 0;
Add support for fallback keycodes. This change enables the framework to synthesize key events to implement default behavior when an application does not handle a key. For example, this change enables numeric keypad keys to perform their associated special function when numlock is off. The application is informed that it is processing a fallback keypress so it can choose to ignore it. Added a new keycode for switching applications. Added ALT key deadkeys. New default key mappings: - ESC -> BACK - Meta+ESC -> HOME - Alt+ESC -> MENU - Meta+Space -> SEARCH - Meta+Tab -> APP_SWITCH Fixed some comments. Fixed some tests. Change-Id: Id7f3b6645f3a350275e624547822f72652f3defe
2010-12-06 17:13:33 -08:00
if (reset || ledState.on != desiredState) {
Add keycodes and meta-key modifiers to support external keyboards. Added new key maps for external keyboards. These maps are intended to be shared across devices by inheriting the "keyboards.mk" product makefile as part of the device's product definition. One of the trickier changes here was to unwind some code in MetaKeyKeyListener that assumed that only the low 8 bits of the meta key state were actually used. The new code abandons bitshifts in favor of simple conditionals that are probably easier to read anyways. The special meta key state constants used by MetaKeyKeyListener are now (@hide) defined in KeyEvent now so as to make it clearer that they share the same code space even if those codes are not valid for KeyEvents. The EventHub now takes care of detecting the appropriate key layout map and key character map when the device is added and sets system properties accordingly. This avoids having duplicate code in KeyCharacterMap to probe for the appropriate key character map although the current probing mechanism has been preserved for legacy reasons just in case. Added support for tracking caps lock, num lock and scroll lock and turning their corresponding LEDs on and off as needed. The key character map format will need to be updated to correctly support PC style external keyboard semantics related to modifier keys. That will come in a later change so caps lock doesn't actually do anything right now except turn the shiny LEDs on and off... Added a list of symbolic key names to KeyEvent and improved the toString() output for debug diagnosis. Having this list in a central place in the framework also allows us to remove it from Monkey so there is one less thing to maintain when we add new keycodes. Bug: 2912307 Change-Id: If8c25e8d50a7c29bbf5d663c94284f5f86de5da4
2010-09-12 17:55:08 -07:00
getEventHub()->setLedState(getDeviceId(), led, desiredState);
ledState.on = desiredState;
}
}
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
// --- CursorInputMapper ---
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
CursorInputMapper::CursorInputMapper(InputDevice* device) :
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
InputMapper(device) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
CursorInputMapper::~CursorInputMapper() {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
uint32_t CursorInputMapper::getSources() {
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
return mSource;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
void CursorInputMapper::populateDeviceInfo(InputDeviceInfo* info) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
InputMapper::populateDeviceInfo(info);
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
if (mParameters.mode == Parameters::MODE_POINTER) {
float minX, minY, maxX, maxY;
if (mPointerController->getBounds(&minX, &minY, &maxX, &maxY)) {
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
info->addMotionRange(AMOTION_EVENT_AXIS_X, mSource, minX, maxX, 0.0f, 0.0f);
info->addMotionRange(AMOTION_EVENT_AXIS_Y, mSource, minY, maxY, 0.0f, 0.0f);
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
}
} else {
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
info->addMotionRange(AMOTION_EVENT_AXIS_X, mSource, -1.0f, 1.0f, 0.0f, mXScale);
info->addMotionRange(AMOTION_EVENT_AXIS_Y, mSource, -1.0f, 1.0f, 0.0f, mYScale);
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
}
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
info->addMotionRange(AMOTION_EVENT_AXIS_PRESSURE, mSource, 0.0f, 1.0f, 0.0f, 0.0f);
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mCursorScrollAccumulator.haveRelativeVWheel()) {
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
info->addMotionRange(AMOTION_EVENT_AXIS_VSCROLL, mSource, -1.0f, 1.0f, 0.0f, 0.0f);
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mCursorScrollAccumulator.haveRelativeHWheel()) {
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
info->addMotionRange(AMOTION_EVENT_AXIS_HSCROLL, mSource, -1.0f, 1.0f, 0.0f, 0.0f);
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
void CursorInputMapper::dump(String8& dump) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.append(INDENT2 "Cursor Input Mapper:\n");
dumpParameters(dump);
dump.appendFormat(INDENT3 "XScale: %0.3f\n", mXScale);
dump.appendFormat(INDENT3 "YScale: %0.3f\n", mYScale);
dump.appendFormat(INDENT3 "XPrecision: %0.3f\n", mXPrecision);
dump.appendFormat(INDENT3 "YPrecision: %0.3f\n", mYPrecision);
dump.appendFormat(INDENT3 "HaveVWheel: %s\n",
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
toString(mCursorScrollAccumulator.haveRelativeVWheel()));
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.appendFormat(INDENT3 "HaveHWheel: %s\n",
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
toString(mCursorScrollAccumulator.haveRelativeHWheel()));
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.appendFormat(INDENT3 "VWheelScale: %0.3f\n", mVWheelScale);
dump.appendFormat(INDENT3 "HWheelScale: %0.3f\n", mHWheelScale);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dump.appendFormat(INDENT3 "Orientation: %d\n", mOrientation);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.appendFormat(INDENT3 "ButtonState: 0x%08x\n", mButtonState);
dump.appendFormat(INDENT3 "Down: %s\n", toString(isPointerDown(mButtonState)));
dump.appendFormat(INDENT3 "DownTime: %lld\n", mDownTime);
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void CursorInputMapper::configure(nsecs_t when,
const InputReaderConfiguration* config, uint32_t changes) {
InputMapper::configure(when, config, changes);
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (!changes) { // first time only
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mCursorScrollAccumulator.configure(getDevice());
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
// Configure basic parameters.
configureParameters();
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
// Configure device mode.
switch (mParameters.mode) {
case Parameters::MODE_POINTER:
mSource = AINPUT_SOURCE_MOUSE;
mXPrecision = 1.0f;
mYPrecision = 1.0f;
mXScale = 1.0f;
mYScale = 1.0f;
mPointerController = getPolicy()->obtainPointerController(getDeviceId());
break;
case Parameters::MODE_NAVIGATION:
mSource = AINPUT_SOURCE_TRACKBALL;
mXPrecision = TRACKBALL_MOVEMENT_THRESHOLD;
mYPrecision = TRACKBALL_MOVEMENT_THRESHOLD;
mXScale = 1.0f / TRACKBALL_MOVEMENT_THRESHOLD;
mYScale = 1.0f / TRACKBALL_MOVEMENT_THRESHOLD;
break;
}
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
mVWheelScale = 1.0f;
mHWheelScale = 1.0f;
}
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (!changes || (changes & InputReaderConfiguration::CHANGE_POINTER_SPEED)) {
mPointerVelocityControl.setParameters(config->pointerVelocityControlParameters);
mWheelXVelocityControl.setParameters(config->wheelVelocityControlParameters);
mWheelYVelocityControl.setParameters(config->wheelVelocityControlParameters);
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (!changes || (changes & InputReaderConfiguration::CHANGE_DISPLAY_INFO)) {
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
if (mParameters.orientationAware && mParameters.hasAssociatedDisplay) {
DisplayViewport v;
if (config->getDisplayInfo(false /*external*/, &v)) {
mOrientation = v.orientation;
} else {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mOrientation = DISPLAY_ORIENTATION_0;
}
} else {
mOrientation = DISPLAY_ORIENTATION_0;
}
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
bumpGeneration();
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
}
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
}
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
void CursorInputMapper::configureParameters() {
mParameters.mode = Parameters::MODE_POINTER;
String8 cursorModeString;
if (getDevice()->getConfiguration().tryGetProperty(String8("cursor.mode"), cursorModeString)) {
if (cursorModeString == "navigation") {
mParameters.mode = Parameters::MODE_NAVIGATION;
} else if (cursorModeString != "pointer" && cursorModeString != "default") {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW("Invalid value for cursor.mode: '%s'", cursorModeString.string());
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
}
}
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
mParameters.orientationAware = false;
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
getDevice()->getConfiguration().tryGetProperty(String8("cursor.orientationAware"),
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
mParameters.orientationAware);
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
mParameters.hasAssociatedDisplay = false;
Report the external display size to the input reader. The input reader needs this information so that it knows how to interpolate touches on an external touch screen. Changed Display so that it asks the WindowManager what the real display size is (as opposed to the raw display size). This means it now takes into the forced display size set by adb shell am display-size. Replaced all calls to getRealWidth() / getRealHeight() / getRealMetrics() in the WindowManager and replaced them with direct usages of the mCurDisplayWidth / mCurDisplayHeight so that the WM doesn't end up making a reentrant Binder call into itself. Fixed the table status bar HeightReceiver so that it updates the height on all configuration changes since it is possible that the display size changed independently of an external HDMI display being plugged / unplugged. Improved the Display class documentation to make the distinctions betweeen the various sizes clearer. Change-Id: I3f75de559d3ebffed532ab46c4ae52c5e7f1da2b
2011-07-25 12:58:12 -07:00
if (mParameters.mode == Parameters::MODE_POINTER || mParameters.orientationAware) {
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
mParameters.hasAssociatedDisplay = true;
Report the external display size to the input reader. The input reader needs this information so that it knows how to interpolate touches on an external touch screen. Changed Display so that it asks the WindowManager what the real display size is (as opposed to the raw display size). This means it now takes into the forced display size set by adb shell am display-size. Replaced all calls to getRealWidth() / getRealHeight() / getRealMetrics() in the WindowManager and replaced them with direct usages of the mCurDisplayWidth / mCurDisplayHeight so that the WM doesn't end up making a reentrant Binder call into itself. Fixed the table status bar HeightReceiver so that it updates the height on all configuration changes since it is possible that the display size changed independently of an external HDMI display being plugged / unplugged. Improved the Display class documentation to make the distinctions betweeen the various sizes clearer. Change-Id: I3f75de559d3ebffed532ab46c4ae52c5e7f1da2b
2011-07-25 12:58:12 -07:00
}
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
}
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
void CursorInputMapper::dumpParameters(String8& dump) {
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
dump.append(INDENT3 "Parameters:\n");
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
dump.appendFormat(INDENT4 "HasAssociatedDisplay: %s\n",
toString(mParameters.hasAssociatedDisplay));
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
switch (mParameters.mode) {
case Parameters::MODE_POINTER:
dump.append(INDENT4 "Mode: pointer\n");
break;
case Parameters::MODE_NAVIGATION:
dump.append(INDENT4 "Mode: navigation\n");
break;
default:
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(false);
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
}
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
dump.appendFormat(INDENT4 "OrientationAware: %s\n",
toString(mParameters.orientationAware));
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void CursorInputMapper::reset(nsecs_t when) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mButtonState = 0;
mDownTime = 0;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mPointerVelocityControl.reset();
mWheelXVelocityControl.reset();
mWheelYVelocityControl.reset();
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mCursorButtonAccumulator.reset(getDevice());
mCursorMotionAccumulator.reset(getDevice());
mCursorScrollAccumulator.reset(getDevice());
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
InputMapper::reset(when);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
void CursorInputMapper::process(const RawEvent* rawEvent) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
mCursorButtonAccumulator.process(rawEvent);
mCursorMotionAccumulator.process(rawEvent);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mCursorScrollAccumulator.process(rawEvent);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
if (rawEvent->type == EV_SYN && rawEvent->code == SYN_REPORT) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
sync(rawEvent->when);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
void CursorInputMapper::sync(nsecs_t when) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
int32_t lastButtonState = mButtonState;
int32_t currentButtonState = mCursorButtonAccumulator.getButtonState();
mButtonState = currentButtonState;
bool wasDown = isPointerDown(lastButtonState);
bool down = isPointerDown(currentButtonState);
bool downChanged;
if (!wasDown && down) {
mDownTime = when;
downChanged = true;
} else if (wasDown && !down) {
downChanged = true;
} else {
downChanged = false;
}
nsecs_t downTime = mDownTime;
bool buttonsChanged = currentButtonState != lastButtonState;
Improve input device wake heuristics. Bug: 5205674 Only wake the device on positive interactions from the user such as button presses, movement, initial touch down events. In particular, do not wake the device on up events since the driver might synthesize them on power off, causing the device to wake up again for no good reason. Change-Id: I767f553ea36d110e6f3a10611b324487ba7d880d
2011-08-23 21:32:42 -07:00
bool buttonsPressed = currentButtonState & ~lastButtonState;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
float deltaX = mCursorMotionAccumulator.getRelativeX() * mXScale;
float deltaY = mCursorMotionAccumulator.getRelativeY() * mYScale;
bool moved = deltaX != 0 || deltaY != 0;
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Rotate delta according to orientation if needed.
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
if (mParameters.orientationAware && mParameters.hasAssociatedDisplay
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
&& (deltaX != 0.0f || deltaY != 0.0f)) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
rotateDelta(mOrientation, &deltaX, &deltaY);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Move the pointer.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
PointerProperties pointerProperties;
pointerProperties.clear();
pointerProperties.id = 0;
pointerProperties.toolType = AMOTION_EVENT_TOOL_TYPE_MOUSE;
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
PointerCoords pointerCoords;
pointerCoords.clear();
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
float vscroll = mCursorScrollAccumulator.getRelativeVWheel();
float hscroll = mCursorScrollAccumulator.getRelativeHWheel();
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
bool scrolled = vscroll != 0 || hscroll != 0;
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mWheelYVelocityControl.move(when, NULL, &vscroll);
mWheelXVelocityControl.move(when, &hscroll, NULL);
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mPointerVelocityControl.move(when, &deltaX, &deltaY);
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (mPointerController != NULL) {
if (moved || scrolled || buttonsChanged) {
mPointerController->setPresentation(
PointerControllerInterface::PRESENTATION_POINTER);
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (moved) {
mPointerController->move(deltaX, deltaY);
}
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (buttonsChanged) {
mPointerController->setButtonState(currentButtonState);
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
}
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mPointerController->unfade(PointerControllerInterface::TRANSITION_IMMEDIATE);
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
float x, y;
mPointerController->getPosition(&x, &y);
pointerCoords.setAxisValue(AMOTION_EVENT_AXIS_X, x);
pointerCoords.setAxisValue(AMOTION_EVENT_AXIS_Y, y);
} else {
pointerCoords.setAxisValue(AMOTION_EVENT_AXIS_X, deltaX);
pointerCoords.setAxisValue(AMOTION_EVENT_AXIS_Y, deltaY);
}
pointerCoords.setAxisValue(AMOTION_EVENT_AXIS_PRESSURE, down ? 1.0f : 0.0f);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Wake screen from external HID peripherals. Added some plumbing to enable the policy to intercept motion events when the screen is off to handle wakeup if needed. Added a basic concept of an external device to limit the scope of the wakeup policy to external devices only. The wakeup policy for internal devices should be based on explicit rules such as policy flags in key layout files. Moved isTouchEvent to native. Ensure the dispatcher sends the right event type to userActivity for non-touch pointer events like HOVER_MOVE and SCROLL. Bug: 3193114 Change-Id: I15dbd48a16810dfaf226ff7ad117d46908ca4f86
2011-03-02 19:23:13 -08:00
// Moving an external trackball or mouse should wake the device.
// We don't do this for internal cursor devices to prevent them from waking up
// the device in your pocket.
// TODO: Use the input device configuration to control this behavior more finely.
uint32_t policyFlags = 0;
Improve input device wake heuristics. Bug: 5205674 Only wake the device on positive interactions from the user such as button presses, movement, initial touch down events. In particular, do not wake the device on up events since the driver might synthesize them on power off, causing the device to wake up again for no good reason. Change-Id: I767f553ea36d110e6f3a10611b324487ba7d880d
2011-08-23 21:32:42 -07:00
if ((buttonsPressed || moved || scrolled) && getDevice()->isExternal()) {
Wake screen from external HID peripherals. Added some plumbing to enable the policy to intercept motion events when the screen is off to handle wakeup if needed. Added a basic concept of an external device to limit the scope of the wakeup policy to external devices only. The wakeup policy for internal devices should be based on explicit rules such as policy flags in key layout files. Moved isTouchEvent to native. Ensure the dispatcher sends the right event type to userActivity for non-touch pointer events like HOVER_MOVE and SCROLL. Bug: 3193114 Change-Id: I15dbd48a16810dfaf226ff7ad117d46908ca4f86
2011-03-02 19:23:13 -08:00
policyFlags |= POLICY_FLAG_WAKE_DROPPED;
}
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
// Synthesize key down from buttons if needed.
synthesizeButtonKeys(getContext(), AKEY_EVENT_ACTION_DOWN, when, getDeviceId(), mSource,
policyFlags, lastButtonState, currentButtonState);
// Send motion event.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (downChanged || moved || scrolled || buttonsChanged) {
int32_t metaState = mContext->getGlobalMetaState();
int32_t motionEventAction;
if (downChanged) {
motionEventAction = down ? AMOTION_EVENT_ACTION_DOWN : AMOTION_EVENT_ACTION_UP;
} else if (down || mPointerController == NULL) {
motionEventAction = AMOTION_EVENT_ACTION_MOVE;
} else {
motionEventAction = AMOTION_EVENT_ACTION_HOVER_MOVE;
}
NotifyMotionArgs args(when, getDeviceId(), mSource, policyFlags,
motionEventAction, 0, metaState, currentButtonState, 0,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
1, &pointerProperties, &pointerCoords, mXPrecision, mYPrecision, downTime);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
getListener()->notifyMotion(&args);
// Send hover move after UP to tell the application that the mouse is hovering now.
if (motionEventAction == AMOTION_EVENT_ACTION_UP
&& mPointerController != NULL) {
NotifyMotionArgs hoverArgs(when, getDeviceId(), mSource, policyFlags,
AMOTION_EVENT_ACTION_HOVER_MOVE, 0,
metaState, currentButtonState, AMOTION_EVENT_EDGE_FLAG_NONE,
1, &pointerProperties, &pointerCoords, mXPrecision, mYPrecision, downTime);
getListener()->notifyMotion(&hoverArgs);
}
Add support for mouse hover and scroll wheel. Dispatch ACTION_HOVER_MOVE and ACTION_SCROLL through the View hierarchy as onGenericTouchEvent. Pointer events dispatched this way are delivered to the view under the pointer. Non-pointer events continue to be delivered to the focused view. Added scroll wheel support to AbsListView, ScrollView, HorizontalScrollView and WebView. Shift+VSCROLL is translated to HSCROLL as appropriate. Added logging of new pointer events in PointerLocationView. Fixed a problem in EventHub when a USB device is removed that resulted in a long stream of ENODEV errors being logged until INotify noticed the device was gone. Note that the new events are not supported by wallpapers at this time because the wallpaper engine only delivers touch events. Make all mouse buttons behave identically. (Effectively we only support one button.) Change-Id: I9ab445ffb63c813fcb07db6693987b02475f3756
2011-02-24 20:55:35 -08:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// Send scroll events.
if (scrolled) {
pointerCoords.setAxisValue(AMOTION_EVENT_AXIS_VSCROLL, vscroll);
pointerCoords.setAxisValue(AMOTION_EVENT_AXIS_HSCROLL, hscroll);
Add support for mouse hover and scroll wheel. Dispatch ACTION_HOVER_MOVE and ACTION_SCROLL through the View hierarchy as onGenericTouchEvent. Pointer events dispatched this way are delivered to the view under the pointer. Non-pointer events continue to be delivered to the focused view. Added scroll wheel support to AbsListView, ScrollView, HorizontalScrollView and WebView. Shift+VSCROLL is translated to HSCROLL as appropriate. Added logging of new pointer events in PointerLocationView. Fixed a problem in EventHub when a USB device is removed that resulted in a long stream of ENODEV errors being logged until INotify noticed the device was gone. Note that the new events are not supported by wallpapers at this time because the wallpaper engine only delivers touch events. Make all mouse buttons behave identically. (Effectively we only support one button.) Change-Id: I9ab445ffb63c813fcb07db6693987b02475f3756
2011-02-24 20:55:35 -08:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
NotifyMotionArgs scrollArgs(when, getDeviceId(), mSource, policyFlags,
AMOTION_EVENT_ACTION_SCROLL, 0, metaState, currentButtonState,
AMOTION_EVENT_EDGE_FLAG_NONE,
1, &pointerProperties, &pointerCoords, mXPrecision, mYPrecision, downTime);
getListener()->notifyMotion(&scrollArgs);
}
Add support for mouse hover and scroll wheel. Dispatch ACTION_HOVER_MOVE and ACTION_SCROLL through the View hierarchy as onGenericTouchEvent. Pointer events dispatched this way are delivered to the view under the pointer. Non-pointer events continue to be delivered to the focused view. Added scroll wheel support to AbsListView, ScrollView, HorizontalScrollView and WebView. Shift+VSCROLL is translated to HSCROLL as appropriate. Added logging of new pointer events in PointerLocationView. Fixed a problem in EventHub when a USB device is removed that resulted in a long stream of ENODEV errors being logged until INotify noticed the device was gone. Note that the new events are not supported by wallpapers at this time because the wallpaper engine only delivers touch events. Make all mouse buttons behave identically. (Effectively we only support one button.) Change-Id: I9ab445ffb63c813fcb07db6693987b02475f3756
2011-02-24 20:55:35 -08:00
}
Add MotionEvent.HOVER_ENTER and HOVER_EXIT. The input dispatcher sends a HOVER_ENTER to a window before dispatching it any HOVER_MOVE events. For compatibility reasons, the window will *also* receive the HOVER_MOVE. When the pointer moves into a different window or the pointer goes down or when events are canceled for some reason, the input dispatcher sends a HOVER_EXIT to the previously hovered window. The view hierarchy behavior is similar. All views under the pointer receive onHoverEvent with HOVER_ENTER followed by any number of HOVER_MOVE events. When the pointer leaves a view, the view receives HOVER_EXIT. Similarly, if a parent view decides to capture hover by returning true from onHoverEvent, the hovered descendants will receive HOVER_EXIT. The default behavior of onHoverEvent is to update the view's hovered state by calling setHovered(true/false). Views can query their current hovered state using isHovered(). For testing purposes, the hovered state is mapped to the pressed drawable state. This will change in a subsequent commit with the introduction of a new hovered drawable state. Change-Id: Ib76a7a90236c8f2c7336e55773acade6346cacbe
2011-03-07 16:56:21 -08:00
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
// Synthesize key up from buttons if needed.
synthesizeButtonKeys(getContext(), AKEY_EVENT_ACTION_UP, when, getDeviceId(), mSource,
policyFlags, lastButtonState, currentButtonState);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mCursorMotionAccumulator.finishSync();
mCursorScrollAccumulator.finishSync();
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
int32_t CursorInputMapper::getScanCodeState(uint32_t sourceMask, int32_t scanCode) {
Fix safe mode and KeyEvent.getMaxKeyCode(). Bug: 2901731 Change-Id: I78617c1b9dee3790fc590e5af4b5083368873184
2010-08-10 15:47:53 -07:00
if (scanCode >= BTN_MOUSE && scanCode < BTN_JOYSTICK) {
return getEventHub()->getScanCodeState(getDeviceId(), scanCode);
} else {
return AKEY_STATE_UNKNOWN;
}
}
Fade out the mouse pointer after inactivity or other events. Fades out the mouse pointer: - after 15 seconds of inactivity normally - after 3 seconds of inactivity in lights out mode - after a non-modifier key down - after a touch down Extended the native Looper to support enqueuing time delayed messages. This is used by the PointerController to control pointer fade timing. Change-Id: I87792fea7dbe2d9376c78cf354fe3189a484d9da
2011-03-02 14:41:58 -08:00
void CursorInputMapper::fadePointer() {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (mPointerController != NULL) {
mPointerController->fade(PointerControllerInterface::TRANSITION_GRADUAL);
}
Fade out the mouse pointer after inactivity or other events. Fades out the mouse pointer: - after 15 seconds of inactivity normally - after 3 seconds of inactivity in lights out mode - after a non-modifier key down - after a touch down Extended the native Looper to support enqueuing time delayed messages. This is used by the PointerController to control pointer fade timing. Change-Id: I87792fea7dbe2d9376c78cf354fe3189a484d9da
2011-03-02 14:41:58 -08:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// --- TouchInputMapper ---
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
TouchInputMapper::TouchInputMapper(InputDevice* device) :
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
InputMapper(device),
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mSource(0), mDeviceMode(DEVICE_MODE_DISABLED),
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mSurfaceOrientation(-1), mSurfaceWidth(-1), mSurfaceHeight(-1) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
TouchInputMapper::~TouchInputMapper() {
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
uint32_t TouchInputMapper::getSources() {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
return mSource;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
void TouchInputMapper::populateDeviceInfo(InputDeviceInfo* info) {
InputMapper::populateDeviceInfo(info);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mDeviceMode != DEVICE_MODE_DISABLED) {
info->addMotionRange(mOrientedRanges.x);
info->addMotionRange(mOrientedRanges.y);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
info->addMotionRange(mOrientedRanges.pressure);
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mOrientedRanges.haveSize) {
info->addMotionRange(mOrientedRanges.size);
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mOrientedRanges.haveTouchSize) {
info->addMotionRange(mOrientedRanges.touchMajor);
info->addMotionRange(mOrientedRanges.touchMinor);
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mOrientedRanges.haveToolSize) {
info->addMotionRange(mOrientedRanges.toolMajor);
info->addMotionRange(mOrientedRanges.toolMinor);
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mOrientedRanges.haveOrientation) {
info->addMotionRange(mOrientedRanges.orientation);
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mOrientedRanges.haveDistance) {
info->addMotionRange(mOrientedRanges.distance);
}
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mOrientedRanges.haveTilt) {
info->addMotionRange(mOrientedRanges.tilt);
}
if (mCursorScrollAccumulator.haveRelativeVWheel()) {
info->addMotionRange(AMOTION_EVENT_AXIS_VSCROLL, mSource, -1.0f, 1.0f, 0.0f, 0.0f);
}
if (mCursorScrollAccumulator.haveRelativeHWheel()) {
info->addMotionRange(AMOTION_EVENT_AXIS_HSCROLL, mSource, -1.0f, 1.0f, 0.0f, 0.0f);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
void TouchInputMapper::dump(String8& dump) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.append(INDENT2 "Touch Input Mapper:\n");
dumpParameters(dump);
dumpVirtualKeys(dump);
dumpRawPointerAxes(dump);
dumpCalibration(dump);
dumpSurface(dump);
dump.appendFormat(INDENT3 "Translation and Scaling Factors:\n");
dump.appendFormat(INDENT4 "XScale: %0.3f\n", mXScale);
dump.appendFormat(INDENT4 "YScale: %0.3f\n", mYScale);
dump.appendFormat(INDENT4 "XPrecision: %0.3f\n", mXPrecision);
dump.appendFormat(INDENT4 "YPrecision: %0.3f\n", mYPrecision);
dump.appendFormat(INDENT4 "GeometricScale: %0.3f\n", mGeometricScale);
dump.appendFormat(INDENT4 "PressureScale: %0.3f\n", mPressureScale);
dump.appendFormat(INDENT4 "SizeScale: %0.3f\n", mSizeScale);
dump.appendFormat(INDENT4 "OrientationScale: %0.3f\n", mOrientationScale);
dump.appendFormat(INDENT4 "DistanceScale: %0.3f\n", mDistanceScale);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dump.appendFormat(INDENT4 "HaveTilt: %s\n", toString(mHaveTilt));
dump.appendFormat(INDENT4 "TiltXCenter: %0.3f\n", mTiltXCenter);
dump.appendFormat(INDENT4 "TiltXScale: %0.3f\n", mTiltXScale);
dump.appendFormat(INDENT4 "TiltYCenter: %0.3f\n", mTiltYCenter);
dump.appendFormat(INDENT4 "TiltYScale: %0.3f\n", mTiltYScale);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.appendFormat(INDENT3 "Last Button State: 0x%08x\n", mLastButtonState);
dump.appendFormat(INDENT3 "Last Raw Touch: pointerCount=%d\n",
mLastRawPointerData.pointerCount);
for (uint32_t i = 0; i < mLastRawPointerData.pointerCount; i++) {
const RawPointerData::Pointer& pointer = mLastRawPointerData.pointers[i];
dump.appendFormat(INDENT4 "[%d]: id=%d, x=%d, y=%d, pressure=%d, "
"touchMajor=%d, touchMinor=%d, toolMajor=%d, toolMinor=%d, "
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
"orientation=%d, tiltX=%d, tiltY=%d, distance=%d, "
"toolType=%d, isHovering=%s\n", i,
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
pointer.id, pointer.x, pointer.y, pointer.pressure,
pointer.touchMajor, pointer.touchMinor,
pointer.toolMajor, pointer.toolMinor,
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
pointer.orientation, pointer.tiltX, pointer.tiltY, pointer.distance,
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
pointer.toolType, toString(pointer.isHovering));
}
dump.appendFormat(INDENT3 "Last Cooked Touch: pointerCount=%d\n",
mLastCookedPointerData.pointerCount);
for (uint32_t i = 0; i < mLastCookedPointerData.pointerCount; i++) {
const PointerProperties& pointerProperties = mLastCookedPointerData.pointerProperties[i];
const PointerCoords& pointerCoords = mLastCookedPointerData.pointerCoords[i];
dump.appendFormat(INDENT4 "[%d]: id=%d, x=%0.3f, y=%0.3f, pressure=%0.3f, "
"touchMajor=%0.3f, touchMinor=%0.3f, toolMajor=%0.3f, toolMinor=%0.3f, "
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
"orientation=%0.3f, tilt=%0.3f, distance=%0.3f, "
"toolType=%d, isHovering=%s\n", i,
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
pointerProperties.id,
pointerCoords.getX(),
pointerCoords.getY(),
pointerCoords.getAxisValue(AMOTION_EVENT_AXIS_PRESSURE),
pointerCoords.getAxisValue(AMOTION_EVENT_AXIS_TOUCH_MAJOR),
pointerCoords.getAxisValue(AMOTION_EVENT_AXIS_TOUCH_MINOR),
pointerCoords.getAxisValue(AMOTION_EVENT_AXIS_TOOL_MAJOR),
pointerCoords.getAxisValue(AMOTION_EVENT_AXIS_TOOL_MINOR),
pointerCoords.getAxisValue(AMOTION_EVENT_AXIS_ORIENTATION),
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
pointerCoords.getAxisValue(AMOTION_EVENT_AXIS_TILT),
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
pointerCoords.getAxisValue(AMOTION_EVENT_AXIS_DISTANCE),
pointerProperties.toolType,
toString(mLastCookedPointerData.isHovering(i)));
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mDeviceMode == DEVICE_MODE_POINTER) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.appendFormat(INDENT3 "Pointer Gesture Detector:\n");
dump.appendFormat(INDENT4 "XMovementScale: %0.3f\n",
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerXMovementScale);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.appendFormat(INDENT4 "YMovementScale: %0.3f\n",
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerYMovementScale);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.appendFormat(INDENT4 "XZoomScale: %0.3f\n",
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerXZoomScale);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.appendFormat(INDENT4 "YZoomScale: %0.3f\n",
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerYZoomScale);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dump.appendFormat(INDENT4 "MaxSwipeWidth: %f\n",
mPointerGestureMaxSwipeWidth);
}
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void TouchInputMapper::configure(nsecs_t when,
const InputReaderConfiguration* config, uint32_t changes) {
InputMapper::configure(when, config, changes);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
mConfig = *config;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (!changes) { // first time only
// Configure basic parameters.
configureParameters();
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Configure common accumulators.
mCursorScrollAccumulator.configure(getDevice());
mTouchButtonAccumulator.configure(getDevice());
Add initial support for cursor-based pointing devices. Some parts stubbed out but you can plug in a mouse and move a green cursor around to interact with the UI. Change-Id: I80d597a7f11d3bd92041890f74b3c77326975e6e
2010-12-23 17:50:18 -08:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
// Configure absolute axis information.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
configureRawPointerAxes();
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
// Prepare input device calibration.
parseCalibration();
resolveCalibration();
}
if (!changes || (changes & InputReaderConfiguration::CHANGE_POINTER_SPEED)) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Update pointer speed.
mPointerVelocityControl.setParameters(mConfig.pointerVelocityControlParameters);
mWheelXVelocityControl.setParameters(mConfig.wheelVelocityControlParameters);
mWheelYVelocityControl.setParameters(mConfig.wheelVelocityControlParameters);
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
bool resetNeeded = false;
if (!changes || (changes & (InputReaderConfiguration::CHANGE_DISPLAY_INFO
Add a "show touches" option for demos and presentations. Bug: 4569045 Change-Id: I8726ea292dd7def790a5e40d7d7e58968974f896
2011-08-26 17:14:14 -07:00
| InputReaderConfiguration::CHANGE_POINTER_GESTURE_ENABLEMENT
| InputReaderConfiguration::CHANGE_SHOW_TOUCHES))) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Configure device sources, surface dimensions, orientation and
// scaling factors.
configureSurface(when, &resetNeeded);
}
if (changes && resetNeeded) {
// Send reset, unless this is the first time the device has been configured,
// in which case the reader will call reset itself after all mappers are ready.
getDevice()->notifyReset(when);
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
void TouchInputMapper::configureParameters() {
Use pointer mode instead of spots for SEMI_MT touch pads. The spot-based presentation relies on being able to accurately locate and distinguish two or more fingers on the touch pad. We can still do gesture recognition without distinct multitouch but it doesn't make sense to show spots because we can't be sure that the spots will appear in the right places. Change-Id: Ia282d77000bf3b50d395273a1f0ac0bbe55b3c6f
2011-06-03 17:06:16 -07:00
// Use the pointer presentation mode for devices that do not support distinct
// multitouch. The spot-based presentation relies on being able to accurately
// locate two or more fingers on the touch pad.
mParameters.gestureMode = getEventHub()->hasInputProperty(getDeviceId(), INPUT_PROP_SEMI_MT)
? Parameters::GESTURE_MODE_POINTER : Parameters::GESTURE_MODE_SPOTS;
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
Touch pad UX improvements. Fade the pointer spot when swiping or performing a freeform gesture. Support configuring the gesture mode in the device IDC file. Added workaround for devices that report individual finger movements one at a time instead of all at once. Bug: 4124987 Change-Id: I44628b00382ad59306e4ec5c4717d69cc6fbebb8
2011-05-25 18:23:38 -07:00
String8 gestureModeString;
if (getDevice()->getConfiguration().tryGetProperty(String8("touch.gestureMode"),
gestureModeString)) {
if (gestureModeString == "pointer") {
mParameters.gestureMode = Parameters::GESTURE_MODE_POINTER;
} else if (gestureModeString == "spots") {
mParameters.gestureMode = Parameters::GESTURE_MODE_SPOTS;
} else if (gestureModeString != "default") {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW("Invalid value for touch.gestureMode: '%s'", gestureModeString.string());
Touch pad UX improvements. Fade the pointer spot when swiping or performing a freeform gesture. Support configuring the gesture mode in the device IDC file. Added workaround for devices that report individual finger movements one at a time instead of all at once. Bug: 4124987 Change-Id: I44628b00382ad59306e4ec5c4717d69cc6fbebb8
2011-05-25 18:23:38 -07:00
}
}
Prioritize INPUT_PROP_DIRECT first. If the device tells us it's really direct, then it's direct. Change-Id: I83875cf47e865ff6f0def149ad0c68db24d9dc4a
2011-08-26 18:38:46 -07:00
if (getEventHub()->hasInputProperty(getDeviceId(), INPUT_PROP_DIRECT)) {
// The device is a touch screen.
mParameters.deviceType = Parameters::DEVICE_TYPE_TOUCH_SCREEN;
} else if (getEventHub()->hasInputProperty(getDeviceId(), INPUT_PROP_POINTER)) {
// The device is a pointing device like a track pad.
mParameters.deviceType = Parameters::DEVICE_TYPE_POINTER;
} else if (getEventHub()->hasRelativeAxis(getDeviceId(), REL_X)
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
|| getEventHub()->hasRelativeAxis(getDeviceId(), REL_Y)) {
// The device is a cursor device with a touch pad attached.
// By default don't use the touch pad to move the pointer.
mParameters.deviceType = Parameters::DEVICE_TYPE_TOUCH_PAD;
} else {
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
// The device is a touch pad of unknown purpose.
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mParameters.deviceType = Parameters::DEVICE_TYPE_POINTER;
}
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
String8 deviceTypeString;
if (getDevice()->getConfiguration().tryGetProperty(String8("touch.deviceType"),
deviceTypeString)) {
Fix crashes caused by some input devices. The touch screen sometimes reports more than 10 pointers even though that's all we asked for. When this happens, we start dropping events with more than 10 pointers. This confuses applications and causes them to crash. Raised the limit to 16 pointers. Bug: 3331247 The default behavior was to identify all touch devices as touch screens. External devices that are plugged in are more likely to be touch pads not attached to a screen. Changed the default to be a touch pad and renamed some internal constants to avoid confusion. A certain mouse happens to also behave like a touch pad. That caused problems because we would see multiple concurrent traces of motion events coming from the same input device so we would batch them up. Added code to ensure that we don't batch events unless they come from the same *source* in addition to coming from the same *device*. Due to batching or misbehaving drivers, it's possible for the set of pointer ids to be different from what we expect when it comes time to split motion events across windows. As a result, we can generate motion events with 0 pointers. When we try to deliver those events, we cause an error in the InputTransport so we tear down the InputChannel and kill the application. Added code to check out assumption about pointer ids and drop the event gracefully instead. Patched up the tests to take into account the change in default behavior for identifying touch screens and touch pads. Change-Id: Ic364bd4cb4cc6335d4a1213a26d6bdadc7e33505
2011-01-25 16:02:22 -08:00
if (deviceTypeString == "touchScreen") {
mParameters.deviceType = Parameters::DEVICE_TYPE_TOUCH_SCREEN;
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
} else if (deviceTypeString == "touchPad") {
mParameters.deviceType = Parameters::DEVICE_TYPE_TOUCH_PAD;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
} else if (deviceTypeString == "pointer") {
mParameters.deviceType = Parameters::DEVICE_TYPE_POINTER;
Touch pad UX improvements. Fade the pointer spot when swiping or performing a freeform gesture. Support configuring the gesture mode in the device IDC file. Added workaround for devices that report individual finger movements one at a time instead of all at once. Bug: 4124987 Change-Id: I44628b00382ad59306e4ec5c4717d69cc6fbebb8
2011-05-25 18:23:38 -07:00
} else if (deviceTypeString != "default") {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW("Invalid value for touch.deviceType: '%s'", deviceTypeString.string());
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
}
}
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
mParameters.orientationAware = mParameters.deviceType == Parameters::DEVICE_TYPE_TOUCH_SCREEN;
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
getDevice()->getConfiguration().tryGetProperty(String8("touch.orientationAware"),
mParameters.orientationAware);
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
mParameters.hasAssociatedDisplay = false;
Report the external display size to the input reader. The input reader needs this information so that it knows how to interpolate touches on an external touch screen. Changed Display so that it asks the WindowManager what the real display size is (as opposed to the raw display size). This means it now takes into the forced display size set by adb shell am display-size. Replaced all calls to getRealWidth() / getRealHeight() / getRealMetrics() in the WindowManager and replaced them with direct usages of the mCurDisplayWidth / mCurDisplayHeight so that the WM doesn't end up making a reentrant Binder call into itself. Fixed the table status bar HeightReceiver so that it updates the height on all configuration changes since it is possible that the display size changed independently of an external HDMI display being plugged / unplugged. Improved the Display class documentation to make the distinctions betweeen the various sizes clearer. Change-Id: I3f75de559d3ebffed532ab46c4ae52c5e7f1da2b
2011-07-25 12:58:12 -07:00
mParameters.associatedDisplayIsExternal = false;
if (mParameters.orientationAware
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
|| mParameters.deviceType == Parameters::DEVICE_TYPE_TOUCH_SCREEN
Report the external display size to the input reader. The input reader needs this information so that it knows how to interpolate touches on an external touch screen. Changed Display so that it asks the WindowManager what the real display size is (as opposed to the raw display size). This means it now takes into the forced display size set by adb shell am display-size. Replaced all calls to getRealWidth() / getRealHeight() / getRealMetrics() in the WindowManager and replaced them with direct usages of the mCurDisplayWidth / mCurDisplayHeight so that the WM doesn't end up making a reentrant Binder call into itself. Fixed the table status bar HeightReceiver so that it updates the height on all configuration changes since it is possible that the display size changed independently of an external HDMI display being plugged / unplugged. Improved the Display class documentation to make the distinctions betweeen the various sizes clearer. Change-Id: I3f75de559d3ebffed532ab46c4ae52c5e7f1da2b
2011-07-25 12:58:12 -07:00
|| mParameters.deviceType == Parameters::DEVICE_TYPE_POINTER) {
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
mParameters.hasAssociatedDisplay = true;
Report the external display size to the input reader. The input reader needs this information so that it knows how to interpolate touches on an external touch screen. Changed Display so that it asks the WindowManager what the real display size is (as opposed to the raw display size). This means it now takes into the forced display size set by adb shell am display-size. Replaced all calls to getRealWidth() / getRealHeight() / getRealMetrics() in the WindowManager and replaced them with direct usages of the mCurDisplayWidth / mCurDisplayHeight so that the WM doesn't end up making a reentrant Binder call into itself. Fixed the table status bar HeightReceiver so that it updates the height on all configuration changes since it is possible that the display size changed independently of an external HDMI display being plugged / unplugged. Improved the Display class documentation to make the distinctions betweeen the various sizes clearer. Change-Id: I3f75de559d3ebffed532ab46c4ae52c5e7f1da2b
2011-07-25 12:58:12 -07:00
mParameters.associatedDisplayIsExternal =
mParameters.deviceType == Parameters::DEVICE_TYPE_TOUCH_SCREEN
&& getDevice()->isExternal();
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
void TouchInputMapper::dumpParameters(String8& dump) {
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
dump.append(INDENT3 "Parameters:\n");
Touch pad UX improvements. Fade the pointer spot when swiping or performing a freeform gesture. Support configuring the gesture mode in the device IDC file. Added workaround for devices that report individual finger movements one at a time instead of all at once. Bug: 4124987 Change-Id: I44628b00382ad59306e4ec5c4717d69cc6fbebb8
2011-05-25 18:23:38 -07:00
switch (mParameters.gestureMode) {
case Parameters::GESTURE_MODE_POINTER:
dump.append(INDENT4 "GestureMode: pointer\n");
break;
case Parameters::GESTURE_MODE_SPOTS:
dump.append(INDENT4 "GestureMode: spots\n");
break;
default:
assert(false);
}
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
switch (mParameters.deviceType) {
case Parameters::DEVICE_TYPE_TOUCH_SCREEN:
dump.append(INDENT4 "DeviceType: touchScreen\n");
break;
case Parameters::DEVICE_TYPE_TOUCH_PAD:
dump.append(INDENT4 "DeviceType: touchPad\n");
break;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
case Parameters::DEVICE_TYPE_POINTER:
dump.append(INDENT4 "DeviceType: pointer\n");
break;
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
default:
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(false);
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
}
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
dump.appendFormat(INDENT4 "AssociatedDisplay: present=%s, isExternal=%s\n",
toString(mParameters.hasAssociatedDisplay),
toString(mParameters.associatedDisplayIsExternal));
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
dump.appendFormat(INDENT4 "OrientationAware: %s\n",
toString(mParameters.orientationAware));
Input dispatcher ANR handling enhancements. This change is essentially a rewrite of the main input dispatcher loop with the target identification folded in. Since the input dispatcher now has all of the window state, it can make better decisions about when to ANR. Added a .5 second deadline for processing app switch keys. This behavior predates Gingerbread but had not previously been ported. Fixed some timing inaccuracies in the ANR accounting that could cause applications to ANR sooner than they should have. Added a mechanism for tracking key and motion events that have been dispatched to a window so that appropriate cancelation events can be synthesized when recovering from ANR. This change helps to keep applications in sync so they don't end up with stuck buttons upon recovery from ANRs. Added more comments to describe the tricky parts of PollLoop. Change-Id: I13dffca27acb436fc383980db536abc4d8b9e6f1
2010-09-08 11:49:43 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void TouchInputMapper::configureRawPointerAxes() {
mRawPointerAxes.clear();
}
void TouchInputMapper::dumpRawPointerAxes(String8& dump) {
dump.append(INDENT3 "Raw Touch Axes:\n");
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.x, "X");
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.y, "Y");
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.pressure, "Pressure");
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.touchMajor, "TouchMajor");
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.touchMinor, "TouchMinor");
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.toolMajor, "ToolMajor");
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.toolMinor, "ToolMinor");
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.orientation, "Orientation");
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.distance, "Distance");
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.tiltX, "TiltX");
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.tiltY, "TiltY");
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.trackingId, "TrackingId");
dumpRawAbsoluteAxisInfo(dump, mRawPointerAxes.slot, "Slot");
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void TouchInputMapper::configureSurface(nsecs_t when, bool* outResetNeeded) {
int32_t oldDeviceMode = mDeviceMode;
// Determine device mode.
if (mParameters.deviceType == Parameters::DEVICE_TYPE_POINTER
&& mConfig.pointerGesturesEnabled) {
mSource = AINPUT_SOURCE_MOUSE;
mDeviceMode = DEVICE_MODE_POINTER;
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
if (hasStylus()) {
mSource |= AINPUT_SOURCE_STYLUS;
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
} else if (mParameters.deviceType == Parameters::DEVICE_TYPE_TOUCH_SCREEN
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
&& mParameters.hasAssociatedDisplay) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mSource = AINPUT_SOURCE_TOUCHSCREEN;
mDeviceMode = DEVICE_MODE_DIRECT;
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
if (hasStylus()) {
mSource |= AINPUT_SOURCE_STYLUS;
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
} else {
mSource = AINPUT_SOURCE_TOUCHPAD;
mDeviceMode = DEVICE_MODE_UNSCALED;
}
Fix off by one errors in touch motion ranges. (DO NOT MERGE) Report inclusive minimum and maximum ranges for all axes including X and Y. Set mouse pointer bounds to 0..width-1, 0..height-1. Rotate touch and mouse positions more carefully, paying attention to the maximum bounds when calculating the complement of an axis. Simplified the InputReader somewhat and removed support for a couple of poorly defined input device configuration parameters. We now assume that the touch device provides useful absolute axis ranges for the X and Y axes since the alternative does not actually make sense. Bug: 3413541 Change-Id: I121d28a125c4f9618cb283dc460d33ff1a907023
2011-03-03 02:09:54 -08:00
// Ensure we have valid X and Y axes.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (!mRawPointerAxes.x.valid || !mRawPointerAxes.y.valid) {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW(INDENT "Touch device '%s' did not report support for X or Y axis! "
Fix off by one errors in touch motion ranges. (DO NOT MERGE) Report inclusive minimum and maximum ranges for all axes including X and Y. Set mouse pointer bounds to 0..width-1, 0..height-1. Rotate touch and mouse positions more carefully, paying attention to the maximum bounds when calculating the complement of an axis. Simplified the InputReader somewhat and removed support for a couple of poorly defined input device configuration parameters. We now assume that the touch device provides useful absolute axis ranges for the X and Y axes since the alternative does not actually make sense. Bug: 3413541 Change-Id: I121d28a125c4f9618cb283dc460d33ff1a907023
2011-03-03 02:09:54 -08:00
"The device will be inoperable.", getDeviceName().string());
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mDeviceMode = DEVICE_MODE_DISABLED;
return;
Fix off by one errors in touch motion ranges. (DO NOT MERGE) Report inclusive minimum and maximum ranges for all axes including X and Y. Set mouse pointer bounds to 0..width-1, 0..height-1. Rotate touch and mouse positions more carefully, paying attention to the maximum bounds when calculating the complement of an axis. Simplified the InputReader somewhat and removed support for a couple of poorly defined input device configuration parameters. We now assume that the touch device provides useful absolute axis ranges for the X and Y axes since the alternative does not actually make sense. Bug: 3413541 Change-Id: I121d28a125c4f9618cb283dc460d33ff1a907023
2011-03-03 02:09:54 -08:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Get associated display dimensions.
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
if (mParameters.hasAssociatedDisplay) {
if (!mConfig.getDisplayInfo(mParameters.associatedDisplayIsExternal,
&mAssociatedDisplayViewport)) {
Rename (IF_)LOGI(_IF) to (IF_)ALOGI(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156801 Bug: 5449033 Change-Id: Ib08fe86d23db91ee153e9f91a99a35c42b9208ea
2012-01-04 20:05:49 +00:00
ALOGI(INDENT "Touch device '%s' could not query the properties of its associated "
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
"display. The device will be inoperable until the display size "
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
"becomes available.",
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
getDeviceName().string());
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mDeviceMode = DEVICE_MODE_DISABLED;
return;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
}
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Configure dimensions.
int32_t width, height, orientation;
if (mDeviceMode == DEVICE_MODE_DIRECT || mDeviceMode == DEVICE_MODE_POINTER) {
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
width = mAssociatedDisplayViewport.logicalRight - mAssociatedDisplayViewport.logicalLeft;
height = mAssociatedDisplayViewport.logicalBottom - mAssociatedDisplayViewport.logicalTop;
if (mAssociatedDisplayViewport.orientation == DISPLAY_ORIENTATION_90
|| mAssociatedDisplayViewport.orientation == DISPLAY_ORIENTATION_270) {
int32_t temp = height;
height = width;
width = temp;
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
orientation = mParameters.orientationAware ?
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
mAssociatedDisplayViewport.orientation : DISPLAY_ORIENTATION_0;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
} else {
width = mRawPointerAxes.x.maxValue - mRawPointerAxes.x.minValue + 1;
height = mRawPointerAxes.y.maxValue - mRawPointerAxes.y.minValue + 1;
orientation = DISPLAY_ORIENTATION_0;
}
// If moving between pointer modes, need to reset some state.
bool deviceModeChanged;
if (mDeviceMode != oldDeviceMode) {
deviceModeChanged = true;
Add a "show touches" option for demos and presentations. Bug: 4569045 Change-Id: I8726ea292dd7def790a5e40d7d7e58968974f896
2011-08-26 17:14:14 -07:00
mOrientedRanges.clear();
}
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
Add a "show touches" option for demos and presentations. Bug: 4569045 Change-Id: I8726ea292dd7def790a5e40d7d7e58968974f896
2011-08-26 17:14:14 -07:00
// Create pointer controller if needed.
if (mDeviceMode == DEVICE_MODE_POINTER ||
(mDeviceMode == DEVICE_MODE_DIRECT && mConfig.showTouches)) {
if (mPointerController == NULL) {
mPointerController = getPolicy()->obtainPointerController(getDeviceId());
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
}
Add a "show touches" option for demos and presentations. Bug: 4569045 Change-Id: I8726ea292dd7def790a5e40d7d7e58968974f896
2011-08-26 17:14:14 -07:00
} else {
mPointerController.clear();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
bool orientationChanged = mSurfaceOrientation != orientation;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
if (orientationChanged) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mSurfaceOrientation = orientation;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
bool sizeChanged = mSurfaceWidth != width || mSurfaceHeight != height;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (sizeChanged || deviceModeChanged) {
Rename (IF_)LOGI(_IF) to (IF_)ALOGI(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156801 Bug: 5449033 Change-Id: Ib08fe86d23db91ee153e9f91a99a35c42b9208ea
2012-01-04 20:05:49 +00:00
ALOGI("Device reconfigured: id=%d, name='%s', surface size is now %dx%d, mode is %d",
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
getDeviceId(), getDeviceName().string(), width, height, mDeviceMode);
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mSurfaceWidth = width;
mSurfaceHeight = height;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
// Configure X and Y factors.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mXScale = float(width) / (mRawPointerAxes.x.maxValue - mRawPointerAxes.x.minValue + 1);
mYScale = float(height) / (mRawPointerAxes.y.maxValue - mRawPointerAxes.y.minValue + 1);
mXPrecision = 1.0f / mXScale;
mYPrecision = 1.0f / mYScale;
Fix off by one errors in touch motion ranges. (DO NOT MERGE) Report inclusive minimum and maximum ranges for all axes including X and Y. Set mouse pointer bounds to 0..width-1, 0..height-1. Rotate touch and mouse positions more carefully, paying attention to the maximum bounds when calculating the complement of an axis. Simplified the InputReader somewhat and removed support for a couple of poorly defined input device configuration parameters. We now assume that the touch device provides useful absolute axis ranges for the X and Y axes since the alternative does not actually make sense. Bug: 3413541 Change-Id: I121d28a125c4f9618cb283dc460d33ff1a907023
2011-03-03 02:09:54 -08:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.x.axis = AMOTION_EVENT_AXIS_X;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mOrientedRanges.x.source = mSource;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.y.axis = AMOTION_EVENT_AXIS_Y;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mOrientedRanges.y.source = mSource;
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
configureVirtualKeys();
Fix native input dispatch in the emulator. Set a default orientation of ROTATION_0. Added some more careful checks based on whether we have valid absolute axis information from the driver. Reset key repeating during configuration changes since the keyboard device may have been removed. Change-Id: I685960828acffcb17595fc5683309e8064a76714
2010-06-29 16:52:21 -07:00
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
// Scale factor for terms that are not oriented in a particular axis.
// If the pixels are square then xScale == yScale otherwise we fake it
// by choosing an average.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mGeometricScale = avg(mXScale, mYScale);
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
// Size of diagonal axis.
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
float diagonalSize = hypotf(width, height);
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
// Size factors.
if (mCalibration.sizeCalibration != Calibration::SIZE_CALIBRATION_NONE) {
if (mRawPointerAxes.touchMajor.valid
&& mRawPointerAxes.touchMajor.maxValue != 0) {
mSizeScale = 1.0f / mRawPointerAxes.touchMajor.maxValue;
} else if (mRawPointerAxes.toolMajor.valid
&& mRawPointerAxes.toolMajor.maxValue != 0) {
mSizeScale = 1.0f / mRawPointerAxes.toolMajor.maxValue;
} else {
mSizeScale = 0.0f;
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.haveTouchSize = true;
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
mOrientedRanges.haveToolSize = true;
mOrientedRanges.haveSize = true;
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.touchMajor.axis = AMOTION_EVENT_AXIS_TOUCH_MAJOR;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mOrientedRanges.touchMajor.source = mSource;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.touchMajor.min = 0;
mOrientedRanges.touchMajor.max = diagonalSize;
mOrientedRanges.touchMajor.flat = 0;
mOrientedRanges.touchMajor.fuzz = 0;
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.touchMinor = mOrientedRanges.touchMajor;
mOrientedRanges.touchMinor.axis = AMOTION_EVENT_AXIS_TOUCH_MINOR;
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.toolMajor.axis = AMOTION_EVENT_AXIS_TOOL_MAJOR;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mOrientedRanges.toolMajor.source = mSource;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.toolMajor.min = 0;
mOrientedRanges.toolMajor.max = diagonalSize;
mOrientedRanges.toolMajor.flat = 0;
mOrientedRanges.toolMajor.fuzz = 0;
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.toolMinor = mOrientedRanges.toolMajor;
mOrientedRanges.toolMinor.axis = AMOTION_EVENT_AXIS_TOOL_MINOR;
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
mOrientedRanges.size.axis = AMOTION_EVENT_AXIS_SIZE;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mOrientedRanges.size.source = mSource;
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
mOrientedRanges.size.min = 0;
mOrientedRanges.size.max = 1.0;
mOrientedRanges.size.flat = 0;
mOrientedRanges.size.fuzz = 0;
} else {
mSizeScale = 0.0f;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
// Pressure factors.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mPressureScale = 0;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mCalibration.pressureCalibration == Calibration::PRESSURE_CALIBRATION_PHYSICAL
|| mCalibration.pressureCalibration
== Calibration::PRESSURE_CALIBRATION_AMPLITUDE) {
if (mCalibration.havePressureScale) {
mPressureScale = mCalibration.pressureScale;
} else if (mRawPointerAxes.pressure.valid
&& mRawPointerAxes.pressure.maxValue != 0) {
mPressureScale = 1.0f / mRawPointerAxes.pressure.maxValue;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mOrientedRanges.pressure.axis = AMOTION_EVENT_AXIS_PRESSURE;
mOrientedRanges.pressure.source = mSource;
mOrientedRanges.pressure.min = 0;
mOrientedRanges.pressure.max = 1.0;
mOrientedRanges.pressure.flat = 0;
mOrientedRanges.pressure.fuzz = 0;
// Tilt
mTiltXCenter = 0;
mTiltXScale = 0;
mTiltYCenter = 0;
mTiltYScale = 0;
mHaveTilt = mRawPointerAxes.tiltX.valid && mRawPointerAxes.tiltY.valid;
if (mHaveTilt) {
mTiltXCenter = avg(mRawPointerAxes.tiltX.minValue,
mRawPointerAxes.tiltX.maxValue);
mTiltYCenter = avg(mRawPointerAxes.tiltY.minValue,
mRawPointerAxes.tiltY.maxValue);
mTiltXScale = M_PI / 180;
mTiltYScale = M_PI / 180;
mOrientedRanges.haveTilt = true;
mOrientedRanges.tilt.axis = AMOTION_EVENT_AXIS_TILT;
mOrientedRanges.tilt.source = mSource;
mOrientedRanges.tilt.min = 0;
mOrientedRanges.tilt.max = M_PI_2;
mOrientedRanges.tilt.flat = 0;
mOrientedRanges.tilt.fuzz = 0;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
// Orientation
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientationScale = 0;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mHaveTilt) {
mOrientedRanges.haveOrientation = true;
mOrientedRanges.orientation.axis = AMOTION_EVENT_AXIS_ORIENTATION;
mOrientedRanges.orientation.source = mSource;
mOrientedRanges.orientation.min = -M_PI;
mOrientedRanges.orientation.max = M_PI;
mOrientedRanges.orientation.flat = 0;
mOrientedRanges.orientation.fuzz = 0;
} else if (mCalibration.orientationCalibration !=
Calibration::ORIENTATION_CALIBRATION_NONE) {
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
if (mCalibration.orientationCalibration
== Calibration::ORIENTATION_CALIBRATION_INTERPOLATED) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mRawPointerAxes.orientation.valid) {
Improve handling of size and orientation. Some touch device drivers report the range of ABS_MT_ORIENTATION as [0,1]. The Linux multitouch protocol spec says that the 0 value should always be considered vertical. Add a new 'box' calibration mode for size to handle the case where width and height are being provided. The 'diameter' calibration mode is inadequate because it drops one of these components. Change-Id: I7bd744fdff95acd4044d295efe8bd7186dbe8061
2012-06-25 17:31:23 -07:00
if (mRawPointerAxes.orientation.maxValue > 0) {
mOrientationScale = M_PI_2 / mRawPointerAxes.orientation.maxValue;
} else if (mRawPointerAxes.orientation.minValue < 0) {
mOrientationScale = -M_PI_2 / mRawPointerAxes.orientation.minValue;
} else {
mOrientationScale = 0;
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
}
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.haveOrientation = true;
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.orientation.axis = AMOTION_EVENT_AXIS_ORIENTATION;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mOrientedRanges.orientation.source = mSource;
mOrientedRanges.orientation.min = -M_PI_2;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.orientation.max = M_PI_2;
mOrientedRanges.orientation.flat = 0;
mOrientedRanges.orientation.fuzz = 0;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
// Distance
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mDistanceScale = 0;
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
if (mCalibration.distanceCalibration != Calibration::DISTANCE_CALIBRATION_NONE) {
if (mCalibration.distanceCalibration
== Calibration::DISTANCE_CALIBRATION_SCALED) {
if (mCalibration.haveDistanceScale) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mDistanceScale = mCalibration.distanceScale;
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
} else {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mDistanceScale = 1.0f;
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
}
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.haveDistance = true;
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.distance.axis = AMOTION_EVENT_AXIS_DISTANCE;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mOrientedRanges.distance.source = mSource;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.distance.min =
mRawPointerAxes.distance.minValue * mDistanceScale;
mOrientedRanges.distance.max =
Distance max set incorrectly The distance max used in the MotionRange is calculated incorrectly. Looks like a copy-and-paste error. Change-Id: I2b6daab088df0fb69e05682b67ca33524ff35987
2012-03-21 14:39:57 +01:00
mRawPointerAxes.distance.maxValue * mDistanceScale;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedRanges.distance.flat = 0;
mOrientedRanges.distance.fuzz =
mRawPointerAxes.distance.fuzz * mDistanceScale;
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (orientationChanged || sizeChanged || deviceModeChanged) {
Fix off by one errors in touch motion ranges. (DO NOT MERGE) Report inclusive minimum and maximum ranges for all axes including X and Y. Set mouse pointer bounds to 0..width-1, 0..height-1. Rotate touch and mouse positions more carefully, paying attention to the maximum bounds when calculating the complement of an axis. Simplified the InputReader somewhat and removed support for a couple of poorly defined input device configuration parameters. We now assume that the touch device provides useful absolute axis ranges for the X and Y axes since the alternative does not actually make sense. Bug: 3413541 Change-Id: I121d28a125c4f9618cb283dc460d33ff1a907023
2011-03-03 02:09:54 -08:00
// Compute oriented surface dimensions, precision, scales and ranges.
// Note that the maximum value reported is an inclusive maximum value so it is one
// unit less than the total width or height of surface.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
switch (mSurfaceOrientation) {
Mouse pointer integration. Added support for loading the pointer icon from a resource. Moved the system server related bits of the input manager out of libui and into libinput since they do not need to be linked into applications. Change-Id: Iec11e0725b3add2b905c51f8ea2c3b4b0d1a2d67
2011-01-02 16:37:43 -08:00
case DISPLAY_ORIENTATION_90:
case DISPLAY_ORIENTATION_270:
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedSurfaceWidth = mSurfaceHeight;
mOrientedSurfaceHeight = mSurfaceWidth;
mOrientedXPrecision = mYPrecision;
mOrientedYPrecision = mXPrecision;
mOrientedRanges.x.min = 0;
mOrientedRanges.x.max = (mRawPointerAxes.y.maxValue - mRawPointerAxes.y.minValue)
* mYScale;
mOrientedRanges.x.flat = 0;
mOrientedRanges.x.fuzz = mYScale;
mOrientedRanges.y.min = 0;
mOrientedRanges.y.max = (mRawPointerAxes.x.maxValue - mRawPointerAxes.x.minValue)
* mXScale;
mOrientedRanges.y.flat = 0;
mOrientedRanges.y.fuzz = mXScale;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
break;
Fix off by one errors in touch motion ranges. (DO NOT MERGE) Report inclusive minimum and maximum ranges for all axes including X and Y. Set mouse pointer bounds to 0..width-1, 0..height-1. Rotate touch and mouse positions more carefully, paying attention to the maximum bounds when calculating the complement of an axis. Simplified the InputReader somewhat and removed support for a couple of poorly defined input device configuration parameters. We now assume that the touch device provides useful absolute axis ranges for the X and Y axes since the alternative does not actually make sense. Bug: 3413541 Change-Id: I121d28a125c4f9618cb283dc460d33ff1a907023
2011-03-03 02:09:54 -08:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
default:
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mOrientedSurfaceWidth = mSurfaceWidth;
mOrientedSurfaceHeight = mSurfaceHeight;
mOrientedXPrecision = mXPrecision;
mOrientedYPrecision = mYPrecision;
mOrientedRanges.x.min = 0;
mOrientedRanges.x.max = (mRawPointerAxes.x.maxValue - mRawPointerAxes.x.minValue)
* mXScale;
mOrientedRanges.x.flat = 0;
mOrientedRanges.x.fuzz = mXScale;
mOrientedRanges.y.min = 0;
mOrientedRanges.y.max = (mRawPointerAxes.y.maxValue - mRawPointerAxes.y.minValue)
* mYScale;
mOrientedRanges.y.flat = 0;
mOrientedRanges.y.fuzz = mYScale;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
break;
Fix native input dispatch in the emulator. Set a default orientation of ROTATION_0. Added some more careful checks based on whether we have valid absolute axis information from the driver. Reset key repeating during configuration changes since the keyboard device may have been removed. Change-Id: I685960828acffcb17595fc5683309e8064a76714
2010-06-29 16:52:21 -07:00
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Compute pointer gesture detection parameters.
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mDeviceMode == DEVICE_MODE_POINTER) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
int32_t rawWidth = mRawPointerAxes.x.maxValue - mRawPointerAxes.x.minValue + 1;
int32_t rawHeight = mRawPointerAxes.y.maxValue - mRawPointerAxes.y.minValue + 1;
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
float rawDiagonal = hypotf(rawWidth, rawHeight);
Make display manager tell input system about viewports. The window manager is no longer responsible for telling the input system about the display viewport. Change-Id: I932882bae55decef55f25093bb2a7ebac1620bb1
2012-09-08 18:05:28 -07:00
float displayDiagonal = hypotf(width, height);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
// Scale movements such that one whole swipe of the touch pad covers a
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
// given area relative to the diagonal size of the display when no acceleration
// is applied.
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Assume that the touch pad has a square aspect ratio such that movements in
// X and Y of the same number of raw units cover the same physical distance.
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerXMovementScale = mConfig.pointerGestureMovementSpeedRatio
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
* displayDiagonal / rawDiagonal;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerYMovementScale = mPointerXMovementScale;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Scale zooms to cover a smaller range of the display than movements do.
// This value determines the area around the pointer that is affected by freeform
// pointer gestures.
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerXZoomScale = mConfig.pointerGestureZoomSpeedRatio
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
* displayDiagonal / rawDiagonal;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerYZoomScale = mPointerXZoomScale;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
// Max width between pointers to detect a swipe gesture is more than some fraction
// of the diagonal axis of the touch pad. Touches that are wider than this are
// translated into freeform gestures.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mPointerGestureMaxSwipeWidth =
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
mConfig.pointerGestureSwipeMaxWidthRatio * rawDiagonal;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
}
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Abort current pointer usages because the state has changed.
abortPointerUsage(when, 0 /*policyFlags*/);
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Inform the dispatcher about the changes.
*outResetNeeded = true;
Notify applications when input devices change. This change allows the InputManager to keep track of what input devices are registered with the system and when they change. It needs to do this so that it can properly clear its cache of input device properties (especially the key map!) when changes occur. Added new API so that applications can register listeners for input device changes. Fixed a minor bug in EventHub where it didn't handle EPOLLHUP properly so it would spam the log about unsupposed epoll events until inotify noticed that the device was gone and removed it. Change-Id: I937d8c601f7185d4299038bce6a2934fe4fdd2b3
2012-04-12 17:32:48 -07:00
bumpGeneration();
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void TouchInputMapper::dumpSurface(String8& dump) {
dump.appendFormat(INDENT3 "SurfaceWidth: %dpx\n", mSurfaceWidth);
dump.appendFormat(INDENT3 "SurfaceHeight: %dpx\n", mSurfaceHeight);
dump.appendFormat(INDENT3 "SurfaceOrientation: %d\n", mSurfaceOrientation);
Input dispatcher ANR handling enhancements. This change is essentially a rewrite of the main input dispatcher loop with the target identification folded in. Since the input dispatcher now has all of the window state, it can make better decisions about when to ANR. Added a .5 second deadline for processing app switch keys. This behavior predates Gingerbread but had not previously been ported. Fixed some timing inaccuracies in the ANR accounting that could cause applications to ANR sooner than they should have. Added a mechanism for tracking key and motion events that have been dispatched to a window so that appropriate cancelation events can be synthesized when recovering from ANR. This change helps to keep applications in sync so they don't end up with stuck buttons upon recovery from ANRs. Added more comments to describe the tricky parts of PollLoop. Change-Id: I13dffca27acb436fc383980db536abc4d8b9e6f1
2010-09-08 11:49:43 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void TouchInputMapper::configureVirtualKeys() {
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Vector<VirtualKeyDefinition> virtualKeyDefinitions;
Improve support for external keyboards. Use Vendor ID, Product ID and optionally the Version to locate keymaps and configuration files for external devices. Moved virtual key definition parsing to native code so that EventHub can identify touch screens with virtual keys and load the appropriate key layout file. Cleaned up a lot of old code in EventHub. Fixed a regression in ViewRoot's fallback event handling. Fixed a minor bug in FileMap that caused it to try to munmap or close invalid handled when released if the attempt to map the file failed. Added a couple of new String8 conveniences for formatting strings. Modified Tokenizer to fall back to open+read when mmap fails since we can't mmap sysfs files as needed to open the virtual key definition files in /sys/board_properties/. Change-Id: I6ca5e5f9547619fd082ddac47e87ce185da69ee6
2010-12-02 13:50:46 -08:00
getEventHub()->getVirtualKeyDefinitions(getDeviceId(), virtualKeyDefinitions);
Fix native input dispatch in the emulator. Set a default orientation of ROTATION_0. Added some more careful checks based on whether we have valid absolute axis information from the driver. Reset key repeating during configuration changes since the keyboard device may have been removed. Change-Id: I685960828acffcb17595fc5683309e8064a76714
2010-06-29 16:52:21 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mVirtualKeys.clear();
Fix native input dispatch in the emulator. Set a default orientation of ROTATION_0. Added some more careful checks based on whether we have valid absolute axis information from the driver. Reset key repeating during configuration changes since the keyboard device may have been removed. Change-Id: I685960828acffcb17595fc5683309e8064a76714
2010-06-29 16:52:21 -07:00
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
if (virtualKeyDefinitions.size() == 0) {
return;
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mVirtualKeys.setCapacity(virtualKeyDefinitions.size());
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
int32_t touchScreenLeft = mRawPointerAxes.x.minValue;
int32_t touchScreenTop = mRawPointerAxes.y.minValue;
int32_t touchScreenWidth = mRawPointerAxes.x.maxValue - mRawPointerAxes.x.minValue + 1;
int32_t touchScreenHeight = mRawPointerAxes.y.maxValue - mRawPointerAxes.y.minValue + 1;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
for (size_t i = 0; i < virtualKeyDefinitions.size(); i++) {
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
const VirtualKeyDefinition& virtualKeyDefinition =
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
virtualKeyDefinitions[i];
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mVirtualKeys.add();
VirtualKey& virtualKey = mVirtualKeys.editTop();
Fix native input dispatch in the emulator. Set a default orientation of ROTATION_0. Added some more careful checks based on whether we have valid absolute axis information from the driver. Reset key repeating during configuration changes since the keyboard device may have been removed. Change-Id: I685960828acffcb17595fc5683309e8064a76714
2010-06-29 16:52:21 -07:00
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
virtualKey.scanCode = virtualKeyDefinition.scanCode;
int32_t keyCode;
uint32_t flags;
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
if (getEventHub()->mapKey(getDeviceId(), virtualKey.scanCode, 0, &keyCode, &flags)) {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW(INDENT "VirtualKey %d: could not obtain key code, ignoring",
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
virtualKey.scanCode);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mVirtualKeys.pop(); // drop the key
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
continue;
}
Fix native input dispatch in the emulator. Set a default orientation of ROTATION_0. Added some more careful checks based on whether we have valid absolute axis information from the driver. Reset key repeating during configuration changes since the keyboard device may have been removed. Change-Id: I685960828acffcb17595fc5683309e8064a76714
2010-06-29 16:52:21 -07:00
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
virtualKey.keyCode = keyCode;
virtualKey.flags = flags;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
// convert the key definition's display coordinates into touch coordinates for a hit box
int32_t halfWidth = virtualKeyDefinition.width / 2;
int32_t halfHeight = virtualKeyDefinition.height / 2;
virtualKey.hitLeft = (virtualKeyDefinition.centerX - halfWidth)
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
* touchScreenWidth / mSurfaceWidth + touchScreenLeft;
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
virtualKey.hitRight= (virtualKeyDefinition.centerX + halfWidth)
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
* touchScreenWidth / mSurfaceWidth + touchScreenLeft;
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
virtualKey.hitTop = (virtualKeyDefinition.centerY - halfHeight)
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
* touchScreenHeight / mSurfaceHeight + touchScreenTop;
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
virtualKey.hitBottom = (virtualKeyDefinition.centerY + halfHeight)
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
* touchScreenHeight / mSurfaceHeight + touchScreenTop;
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
}
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void TouchInputMapper::dumpVirtualKeys(String8& dump) {
if (!mVirtualKeys.isEmpty()) {
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
dump.append(INDENT3 "Virtual Keys:\n");
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
for (size_t i = 0; i < mVirtualKeys.size(); i++) {
const VirtualKey& virtualKey = mVirtualKeys.itemAt(i);
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
dump.appendFormat(INDENT4 "%d: scanCode=%d, keyCode=%d, "
"hitLeft=%d, hitRight=%d, hitTop=%d, hitBottom=%d\n",
i, virtualKey.scanCode, virtualKey.keyCode,
virtualKey.hitLeft, virtualKey.hitRight,
virtualKey.hitTop, virtualKey.hitBottom);
}
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
void TouchInputMapper::parseCalibration() {
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
const PropertyMap& in = getDevice()->getConfiguration();
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Calibration& out = mCalibration;
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
// Size
out.sizeCalibration = Calibration::SIZE_CALIBRATION_DEFAULT;
String8 sizeCalibrationString;
if (in.tryGetProperty(String8("touch.size.calibration"), sizeCalibrationString)) {
if (sizeCalibrationString == "none") {
out.sizeCalibration = Calibration::SIZE_CALIBRATION_NONE;
} else if (sizeCalibrationString == "geometric") {
out.sizeCalibration = Calibration::SIZE_CALIBRATION_GEOMETRIC;
} else if (sizeCalibrationString == "diameter") {
out.sizeCalibration = Calibration::SIZE_CALIBRATION_DIAMETER;
Improve handling of size and orientation. Some touch device drivers report the range of ABS_MT_ORIENTATION as [0,1]. The Linux multitouch protocol spec says that the 0 value should always be considered vertical. Add a new 'box' calibration mode for size to handle the case where width and height are being provided. The 'diameter' calibration mode is inadequate because it drops one of these components. Change-Id: I7bd744fdff95acd4044d295efe8bd7186dbe8061
2012-06-25 17:31:23 -07:00
} else if (sizeCalibrationString == "box") {
out.sizeCalibration = Calibration::SIZE_CALIBRATION_BOX;
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
} else if (sizeCalibrationString == "area") {
out.sizeCalibration = Calibration::SIZE_CALIBRATION_AREA;
} else if (sizeCalibrationString != "default") {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW("Invalid value for touch.size.calibration: '%s'",
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
sizeCalibrationString.string());
}
}
out.haveSizeScale = in.tryGetProperty(String8("touch.size.scale"),
out.sizeScale);
out.haveSizeBias = in.tryGetProperty(String8("touch.size.bias"),
out.sizeBias);
out.haveSizeIsSummed = in.tryGetProperty(String8("touch.size.isSummed"),
out.sizeIsSummed);
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
// Pressure
out.pressureCalibration = Calibration::PRESSURE_CALIBRATION_DEFAULT;
String8 pressureCalibrationString;
Improve input device calibration format. Added support for calibrating touch size for devices that report size as an area measurement rather than as a width. Fixed some bugs. Bug: 3096045 Change-Id: I30a12e73752883516ed054f8af407204bca45814
2010-10-14 21:42:15 -07:00
if (in.tryGetProperty(String8("touch.pressure.calibration"), pressureCalibrationString)) {
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
if (pressureCalibrationString == "none") {
out.pressureCalibration = Calibration::PRESSURE_CALIBRATION_NONE;
} else if (pressureCalibrationString == "physical") {
out.pressureCalibration = Calibration::PRESSURE_CALIBRATION_PHYSICAL;
} else if (pressureCalibrationString == "amplitude") {
out.pressureCalibration = Calibration::PRESSURE_CALIBRATION_AMPLITUDE;
} else if (pressureCalibrationString != "default") {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW("Invalid value for touch.pressure.calibration: '%s'",
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
pressureCalibrationString.string());
}
}
out.havePressureScale = in.tryGetProperty(String8("touch.pressure.scale"),
out.pressureScale);
// Orientation
out.orientationCalibration = Calibration::ORIENTATION_CALIBRATION_DEFAULT;
String8 orientationCalibrationString;
Improve input device calibration format. Added support for calibrating touch size for devices that report size as an area measurement rather than as a width. Fixed some bugs. Bug: 3096045 Change-Id: I30a12e73752883516ed054f8af407204bca45814
2010-10-14 21:42:15 -07:00
if (in.tryGetProperty(String8("touch.orientation.calibration"), orientationCalibrationString)) {
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
if (orientationCalibrationString == "none") {
out.orientationCalibration = Calibration::ORIENTATION_CALIBRATION_NONE;
} else if (orientationCalibrationString == "interpolated") {
out.orientationCalibration = Calibration::ORIENTATION_CALIBRATION_INTERPOLATED;
Add support for detecting finger orientation. Change-Id: I901fc61f7705fcd7950190dc8782bad616032ee2
2011-01-14 19:09:23 -08:00
} else if (orientationCalibrationString == "vector") {
out.orientationCalibration = Calibration::ORIENTATION_CALIBRATION_VECTOR;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
} else if (orientationCalibrationString != "default") {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW("Invalid value for touch.orientation.calibration: '%s'",
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
orientationCalibrationString.string());
}
}
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
// Distance
out.distanceCalibration = Calibration::DISTANCE_CALIBRATION_DEFAULT;
String8 distanceCalibrationString;
if (in.tryGetProperty(String8("touch.distance.calibration"), distanceCalibrationString)) {
if (distanceCalibrationString == "none") {
out.distanceCalibration = Calibration::DISTANCE_CALIBRATION_NONE;
} else if (distanceCalibrationString == "scaled") {
out.distanceCalibration = Calibration::DISTANCE_CALIBRATION_SCALED;
} else if (distanceCalibrationString != "default") {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW("Invalid value for touch.distance.calibration: '%s'",
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
distanceCalibrationString.string());
}
}
out.haveDistanceScale = in.tryGetProperty(String8("touch.distance.scale"),
out.distanceScale);
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
void TouchInputMapper::resolveCalibration() {
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
// Size
if (mRawPointerAxes.touchMajor.valid || mRawPointerAxes.toolMajor.valid) {
if (mCalibration.sizeCalibration == Calibration::SIZE_CALIBRATION_DEFAULT) {
mCalibration.sizeCalibration = Calibration::SIZE_CALIBRATION_GEOMETRIC;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
} else {
mCalibration.sizeCalibration = Calibration::SIZE_CALIBRATION_NONE;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
// Pressure
if (mRawPointerAxes.pressure.valid) {
if (mCalibration.pressureCalibration == Calibration::PRESSURE_CALIBRATION_DEFAULT) {
mCalibration.pressureCalibration = Calibration::PRESSURE_CALIBRATION_PHYSICAL;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
} else {
mCalibration.pressureCalibration = Calibration::PRESSURE_CALIBRATION_NONE;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
// Orientation
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
if (mRawPointerAxes.orientation.valid) {
if (mCalibration.orientationCalibration == Calibration::ORIENTATION_CALIBRATION_DEFAULT) {
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
mCalibration.orientationCalibration = Calibration::ORIENTATION_CALIBRATION_INTERPOLATED;
}
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
} else {
mCalibration.orientationCalibration = Calibration::ORIENTATION_CALIBRATION_NONE;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
// Distance
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
if (mRawPointerAxes.distance.valid) {
if (mCalibration.distanceCalibration == Calibration::DISTANCE_CALIBRATION_DEFAULT) {
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
mCalibration.distanceCalibration = Calibration::DISTANCE_CALIBRATION_SCALED;
}
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
} else {
mCalibration.distanceCalibration = Calibration::DISTANCE_CALIBRATION_NONE;
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
void TouchInputMapper::dumpCalibration(String8& dump) {
dump.append(INDENT3 "Calibration:\n");
Input dispatcher ANR handling enhancements. This change is essentially a rewrite of the main input dispatcher loop with the target identification folded in. Since the input dispatcher now has all of the window state, it can make better decisions about when to ANR. Added a .5 second deadline for processing app switch keys. This behavior predates Gingerbread but had not previously been ported. Fixed some timing inaccuracies in the ANR accounting that could cause applications to ANR sooner than they should have. Added a mechanism for tracking key and motion events that have been dispatched to a window so that appropriate cancelation events can be synthesized when recovering from ANR. This change helps to keep applications in sync so they don't end up with stuck buttons upon recovery from ANRs. Added more comments to describe the tricky parts of PollLoop. Change-Id: I13dffca27acb436fc383980db536abc4d8b9e6f1
2010-09-08 11:49:43 -07:00
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
// Size
switch (mCalibration.sizeCalibration) {
case Calibration::SIZE_CALIBRATION_NONE:
dump.append(INDENT4 "touch.size.calibration: none\n");
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
break;
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
case Calibration::SIZE_CALIBRATION_GEOMETRIC:
dump.append(INDENT4 "touch.size.calibration: geometric\n");
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
break;
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
case Calibration::SIZE_CALIBRATION_DIAMETER:
dump.append(INDENT4 "touch.size.calibration: diameter\n");
Improve input device calibration format. Added support for calibrating touch size for devices that report size as an area measurement rather than as a width. Fixed some bugs. Bug: 3096045 Change-Id: I30a12e73752883516ed054f8af407204bca45814
2010-10-14 21:42:15 -07:00
break;
Improve handling of size and orientation. Some touch device drivers report the range of ABS_MT_ORIENTATION as [0,1]. The Linux multitouch protocol spec says that the 0 value should always be considered vertical. Add a new 'box' calibration mode for size to handle the case where width and height are being provided. The 'diameter' calibration mode is inadequate because it drops one of these components. Change-Id: I7bd744fdff95acd4044d295efe8bd7186dbe8061
2012-06-25 17:31:23 -07:00
case Calibration::SIZE_CALIBRATION_BOX:
dump.append(INDENT4 "touch.size.calibration: box\n");
break;
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
case Calibration::SIZE_CALIBRATION_AREA:
dump.append(INDENT4 "touch.size.calibration: area\n");
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
break;
default:
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(false);
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
if (mCalibration.haveSizeScale) {
dump.appendFormat(INDENT4 "touch.size.scale: %0.3f\n",
mCalibration.sizeScale);
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
if (mCalibration.haveSizeBias) {
dump.appendFormat(INDENT4 "touch.size.bias: %0.3f\n",
mCalibration.sizeBias);
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
if (mCalibration.haveSizeIsSummed) {
dump.appendFormat(INDENT4 "touch.size.isSummed: %s\n",
toString(mCalibration.sizeIsSummed));
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
// Pressure
switch (mCalibration.pressureCalibration) {
case Calibration::PRESSURE_CALIBRATION_NONE:
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
dump.append(INDENT4 "touch.pressure.calibration: none\n");
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
break;
case Calibration::PRESSURE_CALIBRATION_PHYSICAL:
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
dump.append(INDENT4 "touch.pressure.calibration: physical\n");
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
break;
case Calibration::PRESSURE_CALIBRATION_AMPLITUDE:
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
dump.append(INDENT4 "touch.pressure.calibration: amplitude\n");
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
break;
default:
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(false);
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
if (mCalibration.havePressureScale) {
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
dump.appendFormat(INDENT4 "touch.pressure.scale: %0.3f\n",
mCalibration.pressureScale);
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
// Orientation
switch (mCalibration.orientationCalibration) {
case Calibration::ORIENTATION_CALIBRATION_NONE:
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
dump.append(INDENT4 "touch.orientation.calibration: none\n");
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
break;
case Calibration::ORIENTATION_CALIBRATION_INTERPOLATED:
Move verbose logging of device capabilities to dumpsys. Change-Id: I5d55eaebfdf049bd1ac1e7bab2758886cfefc9d7
2010-09-30 14:33:04 -07:00
dump.append(INDENT4 "touch.orientation.calibration: interpolated\n");
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
break;
Add support for detecting finger orientation. Change-Id: I901fc61f7705fcd7950190dc8782bad616032ee2
2011-01-14 19:09:23 -08:00
case Calibration::ORIENTATION_CALIBRATION_VECTOR:
dump.append(INDENT4 "touch.orientation.calibration: vector\n");
break;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
default:
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(false);
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
// Distance
switch (mCalibration.distanceCalibration) {
case Calibration::DISTANCE_CALIBRATION_NONE:
dump.append(INDENT4 "touch.distance.calibration: none\n");
break;
case Calibration::DISTANCE_CALIBRATION_SCALED:
dump.append(INDENT4 "touch.distance.calibration: scaled\n");
break;
default:
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(false);
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
}
if (mCalibration.haveDistanceScale) {
dump.appendFormat(INDENT4 "touch.distance.scale: %0.3f\n",
mCalibration.distanceScale);
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void TouchInputMapper::reset(nsecs_t when) {
mCursorButtonAccumulator.reset(getDevice());
mCursorScrollAccumulator.reset(getDevice());
mTouchButtonAccumulator.reset(getDevice());
mPointerVelocityControl.reset();
mWheelXVelocityControl.reset();
mWheelYVelocityControl.reset();
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentRawPointerData.clear();
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mLastRawPointerData.clear();
mCurrentCookedPointerData.clear();
mLastCookedPointerData.clear();
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentButtonState = 0;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mLastButtonState = 0;
mCurrentRawVScroll = 0;
mCurrentRawHScroll = 0;
mCurrentFingerIdBits.clear();
mLastFingerIdBits.clear();
mCurrentStylusIdBits.clear();
mLastStylusIdBits.clear();
mCurrentMouseIdBits.clear();
mLastMouseIdBits.clear();
mPointerUsage = POINTER_USAGE_NONE;
mSentHoverEnter = false;
mDownTime = 0;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mCurrentVirtualKey.down = false;
mPointerGesture.reset();
mPointerSimple.reset();
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mPointerController != NULL) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mPointerController->fade(PointerControllerInterface::TRANSITION_GRADUAL);
mPointerController->clearSpots();
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
InputMapper::reset(when);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Fix native input dispatch in the emulator. Set a default orientation of ROTATION_0. Added some more careful checks based on whether we have valid absolute axis information from the driver. Reset key repeating during configuration changes since the keyboard device may have been removed. Change-Id: I685960828acffcb17595fc5683309e8064a76714
2010-06-29 16:52:21 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void TouchInputMapper::process(const RawEvent* rawEvent) {
mCursorButtonAccumulator.process(rawEvent);
mCursorScrollAccumulator.process(rawEvent);
mTouchButtonAccumulator.process(rawEvent);
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
if (rawEvent->type == EV_SYN && rawEvent->code == SYN_REPORT) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
sync(rawEvent->when);
}
}
void TouchInputMapper::sync(nsecs_t when) {
// Sync button state.
mCurrentButtonState = mTouchButtonAccumulator.getButtonState()
| mCursorButtonAccumulator.getButtonState();
// Sync scroll state.
mCurrentRawVScroll = mCursorScrollAccumulator.getRelativeVWheel();
mCurrentRawHScroll = mCursorScrollAccumulator.getRelativeHWheel();
mCursorScrollAccumulator.finishSync();
// Sync touch state.
bool havePointerIds = true;
mCurrentRawPointerData.clear();
syncTouch(when, &havePointerIds);
Refactor how timeouts are calculated. Added a timeout mechanism to EventHub and InputReader so that InputMappers can request timeouts to perform delayed processing of input when needed. Change-Id: Iec2045baaf4e67690b15eef3c09a58d5cac76897
2011-03-17 01:34:19 -07:00
#if DEBUG_RAW_EVENTS
if (!havePointerIds) {
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("syncTouch: pointerCount %d -> %d, no pointer ids",
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mLastRawPointerData.pointerCount,
mCurrentRawPointerData.pointerCount);
Refactor how timeouts are calculated. Added a timeout mechanism to EventHub and InputReader so that InputMappers can request timeouts to perform delayed processing of input when needed. Change-Id: Iec2045baaf4e67690b15eef3c09a58d5cac76897
2011-03-17 01:34:19 -07:00
} else {
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("syncTouch: pointerCount %d -> %d, touching ids 0x%08x -> 0x%08x, "
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
"hovering ids 0x%08x -> 0x%08x",
mLastRawPointerData.pointerCount,
mCurrentRawPointerData.pointerCount,
mLastRawPointerData.touchingIdBits.value,
mCurrentRawPointerData.touchingIdBits.value,
mLastRawPointerData.hoveringIdBits.value,
mCurrentRawPointerData.hoveringIdBits.value);
Refactor how timeouts are calculated. Added a timeout mechanism to EventHub and InputReader so that InputMappers can request timeouts to perform delayed processing of input when needed. Change-Id: Iec2045baaf4e67690b15eef3c09a58d5cac76897
2011-03-17 01:34:19 -07:00
}
#endif
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Reset state that we will compute below.
mCurrentFingerIdBits.clear();
mCurrentStylusIdBits.clear();
mCurrentMouseIdBits.clear();
mCurrentCookedPointerData.clear();
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mDeviceMode == DEVICE_MODE_DISABLED) {
// Drop all input if the device is disabled.
mCurrentRawPointerData.clear();
mCurrentButtonState = 0;
} else {
// Preprocess pointer data.
if (!havePointerIds) {
assignPointerIds();
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Handle policy on initial down or hover events.
uint32_t policyFlags = 0;
Improve input device wake heuristics. Bug: 5205674 Only wake the device on positive interactions from the user such as button presses, movement, initial touch down events. In particular, do not wake the device on up events since the driver might synthesize them on power off, causing the device to wake up again for no good reason. Change-Id: I767f553ea36d110e6f3a10611b324487ba7d880d
2011-08-23 21:32:42 -07:00
bool initialDown = mLastRawPointerData.pointerCount == 0
&& mCurrentRawPointerData.pointerCount != 0;
bool buttonsPressed = mCurrentButtonState & ~mLastButtonState;
if (initialDown || buttonsPressed) {
// If this is a touch screen, hide the pointer on an initial down.
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mDeviceMode == DEVICE_MODE_DIRECT) {
getContext()->fadePointer();
}
// Initial downs on external touch devices should wake the device.
// We don't do this for internal touch screens to prevent them from waking
// up in your pocket.
// TODO: Use the input device configuration to control this behavior more finely.
if (getDevice()->isExternal()) {
policyFlags |= POLICY_FLAG_WAKE_DROPPED;
}
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
}
Wake screen from external HID peripherals. Added some plumbing to enable the policy to intercept motion events when the screen is off to handle wakeup if needed. Added a basic concept of an external device to limit the scope of the wakeup policy to external devices only. The wakeup policy for internal devices should be based on explicit rules such as policy flags in key layout files. Moved isTouchEvent to native. Ensure the dispatcher sends the right event type to userActivity for non-touch pointer events like HOVER_MOVE and SCROLL. Bug: 3193114 Change-Id: I15dbd48a16810dfaf226ff7ad117d46908ca4f86
2011-03-02 19:23:13 -08:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Synthesize key down from raw buttons if needed.
synthesizeButtonKeys(getContext(), AKEY_EVENT_ACTION_DOWN, when, getDeviceId(), mSource,
policyFlags, mLastButtonState, mCurrentButtonState);
// Consume raw off-screen touches before cooking pointer data.
// If touches are consumed, subsequent code will not receive any pointer data.
if (consumeRawTouches(when, policyFlags)) {
mCurrentRawPointerData.clear();
Wake screen from external HID peripherals. Added some plumbing to enable the policy to intercept motion events when the screen is off to handle wakeup if needed. Added a basic concept of an external device to limit the scope of the wakeup policy to external devices only. The wakeup policy for internal devices should be based on explicit rules such as policy flags in key layout files. Moved isTouchEvent to native. Ensure the dispatcher sends the right event type to userActivity for non-touch pointer events like HOVER_MOVE and SCROLL. Bug: 3193114 Change-Id: I15dbd48a16810dfaf226ff7ad117d46908ca4f86
2011-03-02 19:23:13 -08:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Cook pointer data. This call populates the mCurrentCookedPointerData structure
// with cooked pointer data that has the same ids and indices as the raw data.
// The following code can use either the raw or cooked data, as needed.
cookPointerData();
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Dispatch the touches either directly or by translation through a pointer on screen.
Add a "show touches" option for demos and presentations. Bug: 4569045 Change-Id: I8726ea292dd7def790a5e40d7d7e58968974f896
2011-08-26 17:14:14 -07:00
if (mDeviceMode == DEVICE_MODE_POINTER) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
for (BitSet32 idBits(mCurrentRawPointerData.touchingIdBits); !idBits.isEmpty(); ) {
uint32_t id = idBits.clearFirstMarkedBit();
const RawPointerData::Pointer& pointer = mCurrentRawPointerData.pointerForId(id);
if (pointer.toolType == AMOTION_EVENT_TOOL_TYPE_STYLUS
|| pointer.toolType == AMOTION_EVENT_TOOL_TYPE_ERASER) {
mCurrentStylusIdBits.markBit(id);
} else if (pointer.toolType == AMOTION_EVENT_TOOL_TYPE_FINGER
|| pointer.toolType == AMOTION_EVENT_TOOL_TYPE_UNKNOWN) {
mCurrentFingerIdBits.markBit(id);
} else if (pointer.toolType == AMOTION_EVENT_TOOL_TYPE_MOUSE) {
mCurrentMouseIdBits.markBit(id);
}
}
for (BitSet32 idBits(mCurrentRawPointerData.hoveringIdBits); !idBits.isEmpty(); ) {
uint32_t id = idBits.clearFirstMarkedBit();
const RawPointerData::Pointer& pointer = mCurrentRawPointerData.pointerForId(id);
if (pointer.toolType == AMOTION_EVENT_TOOL_TYPE_STYLUS
|| pointer.toolType == AMOTION_EVENT_TOOL_TYPE_ERASER) {
mCurrentStylusIdBits.markBit(id);
}
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Stylus takes precedence over all tools, then mouse, then finger.
PointerUsage pointerUsage = mPointerUsage;
if (!mCurrentStylusIdBits.isEmpty()) {
mCurrentMouseIdBits.clear();
mCurrentFingerIdBits.clear();
pointerUsage = POINTER_USAGE_STYLUS;
} else if (!mCurrentMouseIdBits.isEmpty()) {
mCurrentFingerIdBits.clear();
pointerUsage = POINTER_USAGE_MOUSE;
} else if (!mCurrentFingerIdBits.isEmpty() || isPointerDown(mCurrentButtonState)) {
pointerUsage = POINTER_USAGE_GESTURES;
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchPointerUsage(when, policyFlags, pointerUsage);
} else {
Add a "show touches" option for demos and presentations. Bug: 4569045 Change-Id: I8726ea292dd7def790a5e40d7d7e58968974f896
2011-08-26 17:14:14 -07:00
if (mDeviceMode == DEVICE_MODE_DIRECT
&& mConfig.showTouches && mPointerController != NULL) {
mPointerController->setPresentation(PointerControllerInterface::PRESENTATION_SPOT);
mPointerController->fade(PointerControllerInterface::TRANSITION_GRADUAL);
mPointerController->setButtonState(mCurrentButtonState);
mPointerController->setSpots(mCurrentCookedPointerData.pointerCoords,
mCurrentCookedPointerData.idToIndex,
mCurrentCookedPointerData.touchingIdBits);
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchHoverExit(when, policyFlags);
dispatchTouches(when, policyFlags);
dispatchHoverEnterAndMove(when, policyFlags);
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Synthesize key up from raw buttons if needed.
synthesizeButtonKeys(getContext(), AKEY_EVENT_ACTION_UP, when, getDeviceId(), mSource,
policyFlags, mLastButtonState, mCurrentButtonState);
}
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
// Copy current touch to last touch in preparation for the next cycle.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mLastRawPointerData.copyFrom(mCurrentRawPointerData);
mLastCookedPointerData.copyFrom(mCurrentCookedPointerData);
mLastButtonState = mCurrentButtonState;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mLastFingerIdBits = mCurrentFingerIdBits;
mLastStylusIdBits = mCurrentStylusIdBits;
mLastMouseIdBits = mCurrentMouseIdBits;
// Clear some transient state.
mCurrentRawVScroll = 0;
mCurrentRawHScroll = 0;
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
}
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
void TouchInputMapper::timeoutExpired(nsecs_t when) {
Add a "show touches" option for demos and presentations. Bug: 4569045 Change-Id: I8726ea292dd7def790a5e40d7d7e58968974f896
2011-08-26 17:14:14 -07:00
if (mDeviceMode == DEVICE_MODE_POINTER) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mPointerUsage == POINTER_USAGE_GESTURES) {
dispatchPointerGestures(when, 0 /*policyFlags*/, true /*isTimeout*/);
}
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
}
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
bool TouchInputMapper::consumeRawTouches(nsecs_t when, uint32_t policyFlags) {
// Check for release of a virtual key.
if (mCurrentVirtualKey.down) {
if (mCurrentRawPointerData.touchingIdBits.isEmpty()) {
// Pointer went up while virtual key was down.
mCurrentVirtualKey.down = false;
if (!mCurrentVirtualKey.ignored) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#if DEBUG_VIRTUAL_KEYS
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("VirtualKeys: Generating key up: keyCode=%d, scanCode=%d",
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentVirtualKey.keyCode, mCurrentVirtualKey.scanCode);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#endif
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dispatchVirtualKey(when, policyFlags,
AKEY_EVENT_ACTION_UP,
AKEY_EVENT_FLAG_FROM_SYSTEM | AKEY_EVENT_FLAG_VIRTUAL_HARD_KEY);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
return true;
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (mCurrentRawPointerData.touchingIdBits.count() == 1) {
uint32_t id = mCurrentRawPointerData.touchingIdBits.firstMarkedBit();
const RawPointerData::Pointer& pointer = mCurrentRawPointerData.pointerForId(id);
const VirtualKey* virtualKey = findVirtualKeyHit(pointer.x, pointer.y);
if (virtualKey && virtualKey->keyCode == mCurrentVirtualKey.keyCode) {
// Pointer is still within the space of the virtual key.
return true;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// Pointer left virtual key area or another pointer also went down.
// Send key cancellation but do not consume the touch yet.
// This is useful when the user swipes through from the virtual key area
// into the main display surface.
mCurrentVirtualKey.down = false;
if (!mCurrentVirtualKey.ignored) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#if DEBUG_VIRTUAL_KEYS
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("VirtualKeys: Canceling key: keyCode=%d, scanCode=%d",
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentVirtualKey.keyCode, mCurrentVirtualKey.scanCode);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#endif
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dispatchVirtualKey(when, policyFlags,
AKEY_EVENT_ACTION_UP,
AKEY_EVENT_FLAG_FROM_SYSTEM | AKEY_EVENT_FLAG_VIRTUAL_HARD_KEY
| AKEY_EVENT_FLAG_CANCELED);
}
}
if (mLastRawPointerData.touchingIdBits.isEmpty()
&& !mCurrentRawPointerData.touchingIdBits.isEmpty()) {
// Pointer just went down. Check for virtual key press or off-screen touches.
uint32_t id = mCurrentRawPointerData.touchingIdBits.firstMarkedBit();
const RawPointerData::Pointer& pointer = mCurrentRawPointerData.pointerForId(id);
if (!isPointInsideSurface(pointer.x, pointer.y)) {
// If exactly one pointer went down, check for virtual key hit.
// Otherwise we will drop the entire stroke.
if (mCurrentRawPointerData.touchingIdBits.count() == 1) {
const VirtualKey* virtualKey = findVirtualKeyHit(pointer.x, pointer.y);
if (virtualKey) {
mCurrentVirtualKey.down = true;
mCurrentVirtualKey.downTime = when;
mCurrentVirtualKey.keyCode = virtualKey->keyCode;
mCurrentVirtualKey.scanCode = virtualKey->scanCode;
mCurrentVirtualKey.ignored = mContext->shouldDropVirtualKey(
when, getDevice(), virtualKey->keyCode, virtualKey->scanCode);
if (!mCurrentVirtualKey.ignored) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#if DEBUG_VIRTUAL_KEYS
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("VirtualKeys: Generating key down: keyCode=%d, scanCode=%d",
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentVirtualKey.keyCode,
mCurrentVirtualKey.scanCode);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#endif
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
dispatchVirtualKey(when, policyFlags,
AKEY_EVENT_ACTION_DOWN,
AKEY_EVENT_FLAG_FROM_SYSTEM | AKEY_EVENT_FLAG_VIRTUAL_HARD_KEY);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
return true;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Filter virtual keys after touches. Adds a new virtualKeyQuietTimeMillis configuration resource that sets the duration for which virtual keys will be dropped after recent touches on screen. The default value is 0; it is intended to be overridden per device using a resource overlay. This change is designed to help in two cases: 1. Swipes from touchscreen into virtual key area. 2. Accidental taps in virtual key area while using on-screen keyboard. Bug: 3089163 Change-Id: Ib912d4f8a4df9966a39cd537d3ec7c24afab7225
2011-01-18 15:10:10 -08:00
// Disable all virtual key touches that happen within a short time interval of the
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// most recent touch within the screen area. The idea is to filter out stray
// virtual key presses when interacting with the touch screen.
Filter virtual keys after touches. Adds a new virtualKeyQuietTimeMillis configuration resource that sets the duration for which virtual keys will be dropped after recent touches on screen. The default value is 0; it is intended to be overridden per device using a resource overlay. This change is designed to help in two cases: 1. Swipes from touchscreen into virtual key area. 2. Accidental taps in virtual key area while using on-screen keyboard. Bug: 3089163 Change-Id: Ib912d4f8a4df9966a39cd537d3ec7c24afab7225
2011-01-18 15:10:10 -08:00
//
// Problems we're trying to solve:
//
// 1. While scrolling a list or dragging the window shade, the user swipes down into a
// virtual key area that is implemented by a separate touch panel and accidentally
// triggers a virtual key.
//
// 2. While typing in the on screen keyboard, the user taps slightly outside the screen
// area and accidentally triggers a virtual key. This often happens when virtual keys
// are layed out below the screen near to where the on screen keyboard's space bar
// is displayed.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (mConfig.virtualKeyQuietTime > 0 && !mCurrentRawPointerData.touchingIdBits.isEmpty()) {
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
mContext->disableVirtualKeysUntil(when + mConfig.virtualKeyQuietTime);
Filter virtual keys after touches. Adds a new virtualKeyQuietTimeMillis configuration resource that sets the duration for which virtual keys will be dropped after recent touches on screen. The default value is 0; it is intended to be overridden per device using a resource overlay. This change is designed to help in two cases: 1. Swipes from touchscreen into virtual key area. 2. Accidental taps in virtual key area while using on-screen keyboard. Bug: 3089163 Change-Id: Ib912d4f8a4df9966a39cd537d3ec7c24afab7225
2011-01-18 15:10:10 -08:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
return false;
Filter virtual keys after touches. Adds a new virtualKeyQuietTimeMillis configuration resource that sets the duration for which virtual keys will be dropped after recent touches on screen. The default value is 0; it is intended to be overridden per device using a resource overlay. This change is designed to help in two cases: 1. Swipes from touchscreen into virtual key area. 2. Accidental taps in virtual key area while using on-screen keyboard. Bug: 3089163 Change-Id: Ib912d4f8a4df9966a39cd537d3ec7c24afab7225
2011-01-18 15:10:10 -08:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void TouchInputMapper::dispatchVirtualKey(nsecs_t when, uint32_t policyFlags,
int32_t keyEventAction, int32_t keyEventFlags) {
int32_t keyCode = mCurrentVirtualKey.keyCode;
int32_t scanCode = mCurrentVirtualKey.scanCode;
nsecs_t downTime = mCurrentVirtualKey.downTime;
int32_t metaState = mContext->getGlobalMetaState();
policyFlags |= POLICY_FLAG_VIRTUAL;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
NotifyKeyArgs args(when, getDeviceId(), AINPUT_SOURCE_KEYBOARD, policyFlags,
keyEventAction, keyEventFlags, keyCode, scanCode, metaState, downTime);
getListener()->notifyKey(&args);
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void TouchInputMapper::dispatchTouches(nsecs_t when, uint32_t policyFlags) {
BitSet32 currentIdBits = mCurrentCookedPointerData.touchingIdBits;
BitSet32 lastIdBits = mLastCookedPointerData.touchingIdBits;
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
int32_t metaState = getContext()->getGlobalMetaState();
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
int32_t buttonState = mCurrentButtonState;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
if (currentIdBits == lastIdBits) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (!currentIdBits.isEmpty()) {
// No pointer id changes so this is a move event.
// The listener takes care of batching moves so we don't have to deal with that here.
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchMotion(when, policyFlags, mSource,
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
AMOTION_EVENT_ACTION_MOVE, 0, metaState, buttonState,
AMOTION_EVENT_EDGE_FLAG_NONE,
mCurrentCookedPointerData.pointerProperties,
mCurrentCookedPointerData.pointerCoords,
mCurrentCookedPointerData.idToIndex,
currentIdBits, -1,
mOrientedXPrecision, mOrientedYPrecision, mDownTime);
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
} else {
Add unit tests for native input and fix bugs identified. Fixed a bug where we would lose the first touch point when swiping out of the virtual key area. Fixed a bug where we would not send an ACTION_MOVE event in cases where individual pointers went down/up and the remaining pointers actually moved. This is important since many applications do not handle pointer movements during ACTION_POINTER_DOWN or ACTION_POINTER_UP. In the case of ACTION_POINTER_UP the movement was completely lost since all pointers were dispatched using their old location rather than the new location. Improved motion event validation to check for duplicate pointer ids. Added an input source constant that was missing from the NDK api but defined in the framework api. Added a timestamp when reporting added/removed devices in EventHub. Bug: 3070082 Change-Id: I3206a030f43b7616e2f48006e5a9d522c4d92e56
2010-10-20 15:33:38 -07:00
// There may be pointers going up and pointers going down and pointers moving
// all at the same time.
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
BitSet32 upIdBits(lastIdBits.value & ~currentIdBits.value);
BitSet32 downIdBits(currentIdBits.value & ~lastIdBits.value);
Add unit tests for native input and fix bugs identified. Fixed a bug where we would lose the first touch point when swiping out of the virtual key area. Fixed a bug where we would not send an ACTION_MOVE event in cases where individual pointers went down/up and the remaining pointers actually moved. This is important since many applications do not handle pointer movements during ACTION_POINTER_DOWN or ACTION_POINTER_UP. In the case of ACTION_POINTER_UP the movement was completely lost since all pointers were dispatched using their old location rather than the new location. Improved motion event validation to check for duplicate pointer ids. Added an input source constant that was missing from the NDK api but defined in the framework api. Added a timestamp when reporting added/removed devices in EventHub. Bug: 3070082 Change-Id: I3206a030f43b7616e2f48006e5a9d522c4d92e56
2010-10-20 15:33:38 -07:00
BitSet32 moveIdBits(lastIdBits.value & currentIdBits.value);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
BitSet32 dispatchedIdBits(lastIdBits.value);
Add unit tests for native input and fix bugs identified. Fixed a bug where we would lose the first touch point when swiping out of the virtual key area. Fixed a bug where we would not send an ACTION_MOVE event in cases where individual pointers went down/up and the remaining pointers actually moved. This is important since many applications do not handle pointer movements during ACTION_POINTER_DOWN or ACTION_POINTER_UP. In the case of ACTION_POINTER_UP the movement was completely lost since all pointers were dispatched using their old location rather than the new location. Improved motion event validation to check for duplicate pointer ids. Added an input source constant that was missing from the NDK api but defined in the framework api. Added a timestamp when reporting added/removed devices in EventHub. Bug: 3070082 Change-Id: I3206a030f43b7616e2f48006e5a9d522c4d92e56
2010-10-20 15:33:38 -07:00
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Update last coordinates of pointers that have moved so that we observe the new
// pointer positions at the same time as other pointers that have just gone up.
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
bool moveNeeded = updateMovedPointers(
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentCookedPointerData.pointerProperties,
mCurrentCookedPointerData.pointerCoords,
mCurrentCookedPointerData.idToIndex,
mLastCookedPointerData.pointerProperties,
mLastCookedPointerData.pointerCoords,
mLastCookedPointerData.idToIndex,
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
moveIdBits);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (buttonState != mLastButtonState) {
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
moveNeeded = true;
}
Add unit tests for native input and fix bugs identified. Fixed a bug where we would lose the first touch point when swiping out of the virtual key area. Fixed a bug where we would not send an ACTION_MOVE event in cases where individual pointers went down/up and the remaining pointers actually moved. This is important since many applications do not handle pointer movements during ACTION_POINTER_DOWN or ACTION_POINTER_UP. In the case of ACTION_POINTER_UP the movement was completely lost since all pointers were dispatched using their old location rather than the new location. Improved motion event validation to check for duplicate pointer ids. Added an input source constant that was missing from the NDK api but defined in the framework api. Added a timestamp when reporting added/removed devices in EventHub. Bug: 3070082 Change-Id: I3206a030f43b7616e2f48006e5a9d522c4d92e56
2010-10-20 15:33:38 -07:00
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Dispatch pointer up events.
Add unit tests for native input and fix bugs identified. Fixed a bug where we would lose the first touch point when swiping out of the virtual key area. Fixed a bug where we would not send an ACTION_MOVE event in cases where individual pointers went down/up and the remaining pointers actually moved. This is important since many applications do not handle pointer movements during ACTION_POINTER_DOWN or ACTION_POINTER_UP. In the case of ACTION_POINTER_UP the movement was completely lost since all pointers were dispatched using their old location rather than the new location. Improved motion event validation to check for duplicate pointer ids. Added an input source constant that was missing from the NDK api but defined in the framework api. Added a timestamp when reporting added/removed devices in EventHub. Bug: 3070082 Change-Id: I3206a030f43b7616e2f48006e5a9d522c4d92e56
2010-10-20 15:33:38 -07:00
while (!upIdBits.isEmpty()) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t upId = upIdBits.clearFirstMarkedBit();
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchMotion(when, policyFlags, mSource,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
AMOTION_EVENT_ACTION_POINTER_UP, 0, metaState, buttonState, 0,
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mLastCookedPointerData.pointerProperties,
mLastCookedPointerData.pointerCoords,
mLastCookedPointerData.idToIndex,
dispatchedIdBits, upId,
mOrientedXPrecision, mOrientedYPrecision, mDownTime);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
dispatchedIdBits.clearBit(upId);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Add unit tests for native input and fix bugs identified. Fixed a bug where we would lose the first touch point when swiping out of the virtual key area. Fixed a bug where we would not send an ACTION_MOVE event in cases where individual pointers went down/up and the remaining pointers actually moved. This is important since many applications do not handle pointer movements during ACTION_POINTER_DOWN or ACTION_POINTER_UP. In the case of ACTION_POINTER_UP the movement was completely lost since all pointers were dispatched using their old location rather than the new location. Improved motion event validation to check for duplicate pointer ids. Added an input source constant that was missing from the NDK api but defined in the framework api. Added a timestamp when reporting added/removed devices in EventHub. Bug: 3070082 Change-Id: I3206a030f43b7616e2f48006e5a9d522c4d92e56
2010-10-20 15:33:38 -07:00
// Dispatch move events if any of the remaining pointers moved from their old locations.
// Although applications receive new locations as part of individual pointer up
// events, they do not generally handle them except when presented in a move event.
if (moveNeeded) {
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(moveIdBits.value == dispatchedIdBits.value);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchMotion(when, policyFlags, mSource,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
AMOTION_EVENT_ACTION_MOVE, 0, metaState, buttonState, 0,
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentCookedPointerData.pointerProperties,
mCurrentCookedPointerData.pointerCoords,
mCurrentCookedPointerData.idToIndex,
dispatchedIdBits, -1,
mOrientedXPrecision, mOrientedYPrecision, mDownTime);
Add unit tests for native input and fix bugs identified. Fixed a bug where we would lose the first touch point when swiping out of the virtual key area. Fixed a bug where we would not send an ACTION_MOVE event in cases where individual pointers went down/up and the remaining pointers actually moved. This is important since many applications do not handle pointer movements during ACTION_POINTER_DOWN or ACTION_POINTER_UP. In the case of ACTION_POINTER_UP the movement was completely lost since all pointers were dispatched using their old location rather than the new location. Improved motion event validation to check for duplicate pointer ids. Added an input source constant that was missing from the NDK api but defined in the framework api. Added a timestamp when reporting added/removed devices in EventHub. Bug: 3070082 Change-Id: I3206a030f43b7616e2f48006e5a9d522c4d92e56
2010-10-20 15:33:38 -07:00
}
// Dispatch pointer down events using the new pointer locations.
while (!downIdBits.isEmpty()) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t downId = downIdBits.clearFirstMarkedBit();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
dispatchedIdBits.markBit(downId);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
if (dispatchedIdBits.count() == 1) {
// First pointer is going down. Set down time.
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
mDownTime = when;
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchMotion(when, policyFlags, mSource,
Remove edge slop support. The edge slop feature was added a long time ago for a specific device that had severe non-linear touch response issues near the edges of its touch screen. The feature is not useful on modern hardware that is property calibrated. Change-Id: I26716cbc7411c31883ddb4bb4ddd981027790d0f
2011-07-14 21:48:23 -07:00
AMOTION_EVENT_ACTION_POINTER_DOWN, 0, metaState, buttonState, 0,
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentCookedPointerData.pointerProperties,
mCurrentCookedPointerData.pointerCoords,
mCurrentCookedPointerData.idToIndex,
dispatchedIdBits, downId,
mOrientedXPrecision, mOrientedYPrecision, mDownTime);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void TouchInputMapper::dispatchHoverExit(nsecs_t when, uint32_t policyFlags) {
if (mSentHoverEnter &&
(mCurrentCookedPointerData.hoveringIdBits.isEmpty()
|| !mCurrentCookedPointerData.touchingIdBits.isEmpty())) {
int32_t metaState = getContext()->getGlobalMetaState();
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchMotion(when, policyFlags, mSource,
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
AMOTION_EVENT_ACTION_HOVER_EXIT, 0, metaState, mLastButtonState, 0,
mLastCookedPointerData.pointerProperties,
mLastCookedPointerData.pointerCoords,
mLastCookedPointerData.idToIndex,
mLastCookedPointerData.hoveringIdBits, -1,
mOrientedXPrecision, mOrientedYPrecision, mDownTime);
mSentHoverEnter = false;
}
}
void TouchInputMapper::dispatchHoverEnterAndMove(nsecs_t when, uint32_t policyFlags) {
if (mCurrentCookedPointerData.touchingIdBits.isEmpty()
&& !mCurrentCookedPointerData.hoveringIdBits.isEmpty()) {
int32_t metaState = getContext()->getGlobalMetaState();
if (!mSentHoverEnter) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchMotion(when, policyFlags, mSource,
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
AMOTION_EVENT_ACTION_HOVER_ENTER, 0, metaState, mCurrentButtonState, 0,
mCurrentCookedPointerData.pointerProperties,
mCurrentCookedPointerData.pointerCoords,
mCurrentCookedPointerData.idToIndex,
mCurrentCookedPointerData.hoveringIdBits, -1,
mOrientedXPrecision, mOrientedYPrecision, mDownTime);
mSentHoverEnter = true;
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchMotion(when, policyFlags, mSource,
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
AMOTION_EVENT_ACTION_HOVER_MOVE, 0, metaState, mCurrentButtonState, 0,
mCurrentCookedPointerData.pointerProperties,
mCurrentCookedPointerData.pointerCoords,
mCurrentCookedPointerData.idToIndex,
mCurrentCookedPointerData.hoveringIdBits, -1,
mOrientedXPrecision, mOrientedYPrecision, mDownTime);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
}
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void TouchInputMapper::cookPointerData() {
uint32_t currentPointerCount = mCurrentRawPointerData.pointerCount;
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentCookedPointerData.clear();
mCurrentCookedPointerData.pointerCount = currentPointerCount;
mCurrentCookedPointerData.hoveringIdBits = mCurrentRawPointerData.hoveringIdBits;
mCurrentCookedPointerData.touchingIdBits = mCurrentRawPointerData.touchingIdBits;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// Walk through the the active pointers and map device coordinates onto
// surface coordinates and adjust for display orientation.
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
for (uint32_t i = 0; i < currentPointerCount; i++) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
const RawPointerData::Pointer& in = mCurrentRawPointerData.pointers[i];
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
// Size
float touchMajor, touchMinor, toolMajor, toolMinor, size;
switch (mCalibration.sizeCalibration) {
case Calibration::SIZE_CALIBRATION_GEOMETRIC:
case Calibration::SIZE_CALIBRATION_DIAMETER:
Improve handling of size and orientation. Some touch device drivers report the range of ABS_MT_ORIENTATION as [0,1]. The Linux multitouch protocol spec says that the 0 value should always be considered vertical. Add a new 'box' calibration mode for size to handle the case where width and height are being provided. The 'diameter' calibration mode is inadequate because it drops one of these components. Change-Id: I7bd744fdff95acd4044d295efe8bd7186dbe8061
2012-06-25 17:31:23 -07:00
case Calibration::SIZE_CALIBRATION_BOX:
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
case Calibration::SIZE_CALIBRATION_AREA:
if (mRawPointerAxes.touchMajor.valid && mRawPointerAxes.toolMajor.valid) {
touchMajor = in.touchMajor;
touchMinor = mRawPointerAxes.touchMinor.valid ? in.touchMinor : in.touchMajor;
toolMajor = in.toolMajor;
toolMinor = mRawPointerAxes.toolMinor.valid ? in.toolMinor : in.toolMajor;
size = mRawPointerAxes.touchMinor.valid
? avg(in.touchMajor, in.touchMinor) : in.touchMajor;
} else if (mRawPointerAxes.touchMajor.valid) {
toolMajor = touchMajor = in.touchMajor;
toolMinor = touchMinor = mRawPointerAxes.touchMinor.valid
? in.touchMinor : in.touchMajor;
size = mRawPointerAxes.touchMinor.valid
? avg(in.touchMajor, in.touchMinor) : in.touchMajor;
} else if (mRawPointerAxes.toolMajor.valid) {
touchMajor = toolMajor = in.toolMajor;
touchMinor = toolMinor = mRawPointerAxes.toolMinor.valid
? in.toolMinor : in.toolMajor;
size = mRawPointerAxes.toolMinor.valid
? avg(in.toolMajor, in.toolMinor) : in.toolMajor;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
} else {
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(false, "No touch or tool axes. "
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
"Size calibration should have been resolved to NONE.");
touchMajor = 0;
touchMinor = 0;
toolMajor = 0;
toolMinor = 0;
size = 0;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
if (mCalibration.haveSizeIsSummed && mCalibration.sizeIsSummed) {
uint32_t touchingCount = mCurrentRawPointerData.touchingIdBits.count();
if (touchingCount > 1) {
touchMajor /= touchingCount;
touchMinor /= touchingCount;
toolMajor /= touchingCount;
toolMinor /= touchingCount;
size /= touchingCount;
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
if (mCalibration.sizeCalibration == Calibration::SIZE_CALIBRATION_GEOMETRIC) {
touchMajor *= mGeometricScale;
touchMinor *= mGeometricScale;
toolMajor *= mGeometricScale;
toolMinor *= mGeometricScale;
} else if (mCalibration.sizeCalibration == Calibration::SIZE_CALIBRATION_AREA) {
touchMajor = touchMajor > 0 ? sqrtf(touchMajor) : 0;
touchMinor = touchMajor;
toolMajor = toolMajor > 0 ? sqrtf(toolMajor) : 0;
toolMinor = toolMajor;
} else if (mCalibration.sizeCalibration == Calibration::SIZE_CALIBRATION_DIAMETER) {
touchMinor = touchMajor;
toolMinor = toolMajor;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
mCalibration.applySizeScaleAndBias(&touchMajor);
mCalibration.applySizeScaleAndBias(&touchMinor);
mCalibration.applySizeScaleAndBias(&toolMajor);
mCalibration.applySizeScaleAndBias(&toolMinor);
size *= mSizeScale;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
break;
default:
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
touchMajor = 0;
touchMinor = 0;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
toolMajor = 0;
toolMinor = 0;
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
size = 0;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
break;
}
// Pressure
float pressure;
switch (mCalibration.pressureCalibration) {
case Calibration::PRESSURE_CALIBRATION_PHYSICAL:
case Calibration::PRESSURE_CALIBRATION_AMPLITUDE:
Improve touch device configuration. Bug: 5064702 This change improves compatibility with standard touch device drivers. All existing touch device IDC files will need to be updated because the input system now assumes that the touch device drivers implement the multitouch input protocol correctly and use the appropriate axes. Change-Id: I93aad0e011efea74fbc7fa3da31f7967aff8136c
2011-08-11 00:05:01 -07:00
pressure = in.pressure * mPressureScale;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
break;
default:
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
pressure = in.isHovering ? 0 : 1;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
break;
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
// Tilt and Orientation
float tilt;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
float orientation;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mHaveTilt) {
float tiltXAngle = (in.tiltX - mTiltXCenter) * mTiltXScale;
float tiltYAngle = (in.tiltY - mTiltYCenter) * mTiltYScale;
orientation = atan2f(-sinf(tiltXAngle), sinf(tiltYAngle));
tilt = acosf(cosf(tiltXAngle) * cosf(tiltYAngle));
} else {
tilt = 0;
switch (mCalibration.orientationCalibration) {
case Calibration::ORIENTATION_CALIBRATION_INTERPOLATED:
Improve handling of size and orientation. Some touch device drivers report the range of ABS_MT_ORIENTATION as [0,1]. The Linux multitouch protocol spec says that the 0 value should always be considered vertical. Add a new 'box' calibration mode for size to handle the case where width and height are being provided. The 'diameter' calibration mode is inadequate because it drops one of these components. Change-Id: I7bd744fdff95acd4044d295efe8bd7186dbe8061
2012-06-25 17:31:23 -07:00
orientation = in.orientation * mOrientationScale;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
break;
case Calibration::ORIENTATION_CALIBRATION_VECTOR: {
int32_t c1 = signExtendNybble((in.orientation & 0xf0) >> 4);
int32_t c2 = signExtendNybble(in.orientation & 0x0f);
if (c1 != 0 || c2 != 0) {
orientation = atan2f(c1, c2) * 0.5f;
float confidence = hypotf(c1, c2);
float scale = 1.0f + confidence / 16.0f;
touchMajor *= scale;
touchMinor /= scale;
toolMajor *= scale;
toolMinor /= scale;
} else {
orientation = 0;
}
break;
}
default:
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
orientation = 0;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
// Distance
float distance;
switch (mCalibration.distanceCalibration) {
case Calibration::DISTANCE_CALIBRATION_SCALED:
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
distance = in.distance * mDistanceScale;
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
break;
default:
distance = 0;
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// X and Y
// Adjust coords for surface orientation.
float x, y;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
switch (mSurfaceOrientation) {
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
case DISPLAY_ORIENTATION_90:
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
x = float(in.y - mRawPointerAxes.y.minValue) * mYScale;
y = float(mRawPointerAxes.x.maxValue - in.x) * mXScale;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
orientation -= M_PI_2;
if (orientation < - M_PI_2) {
orientation += M_PI;
}
break;
case DISPLAY_ORIENTATION_180:
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
x = float(mRawPointerAxes.x.maxValue - in.x) * mXScale;
y = float(mRawPointerAxes.y.maxValue - in.y) * mYScale;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
break;
case DISPLAY_ORIENTATION_270:
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
x = float(mRawPointerAxes.y.maxValue - in.y) * mYScale;
y = float(in.x - mRawPointerAxes.x.minValue) * mXScale;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
orientation += M_PI_2;
if (orientation > M_PI_2) {
orientation -= M_PI;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
break;
default:
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
x = float(in.x - mRawPointerAxes.x.minValue) * mXScale;
y = float(in.y - mRawPointerAxes.y.minValue) * mYScale;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
break;
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Write output coords.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
PointerCoords& out = mCurrentCookedPointerData.pointerCoords[i];
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
out.clear();
out.setAxisValue(AMOTION_EVENT_AXIS_X, x);
out.setAxisValue(AMOTION_EVENT_AXIS_Y, y);
out.setAxisValue(AMOTION_EVENT_AXIS_PRESSURE, pressure);
out.setAxisValue(AMOTION_EVENT_AXIS_SIZE, size);
out.setAxisValue(AMOTION_EVENT_AXIS_TOUCH_MAJOR, touchMajor);
out.setAxisValue(AMOTION_EVENT_AXIS_TOUCH_MINOR, touchMinor);
out.setAxisValue(AMOTION_EVENT_AXIS_TOOL_MAJOR, toolMajor);
out.setAxisValue(AMOTION_EVENT_AXIS_TOOL_MINOR, toolMinor);
out.setAxisValue(AMOTION_EVENT_AXIS_ORIENTATION, orientation);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
out.setAxisValue(AMOTION_EVENT_AXIS_TILT, tilt);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
out.setAxisValue(AMOTION_EVENT_AXIS_DISTANCE, distance);
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
// Write output properties.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
PointerProperties& properties = mCurrentCookedPointerData.pointerProperties[i];
uint32_t id = in.id;
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
properties.clear();
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
properties.id = id;
properties.toolType = in.toolType;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// Write id index.
mCurrentCookedPointerData.idToIndex[id] = i;
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void TouchInputMapper::dispatchPointerUsage(nsecs_t when, uint32_t policyFlags,
PointerUsage pointerUsage) {
if (pointerUsage != mPointerUsage) {
abortPointerUsage(when, policyFlags);
mPointerUsage = pointerUsage;
}
switch (mPointerUsage) {
case POINTER_USAGE_GESTURES:
dispatchPointerGestures(when, policyFlags, false /*isTimeout*/);
break;
case POINTER_USAGE_STYLUS:
dispatchPointerStylus(when, policyFlags);
break;
case POINTER_USAGE_MOUSE:
dispatchPointerMouse(when, policyFlags);
break;
default:
break;
}
}
void TouchInputMapper::abortPointerUsage(nsecs_t when, uint32_t policyFlags) {
switch (mPointerUsage) {
case POINTER_USAGE_GESTURES:
abortPointerGestures(when, policyFlags);
break;
case POINTER_USAGE_STYLUS:
abortPointerStylus(when, policyFlags);
break;
case POINTER_USAGE_MOUSE:
abortPointerMouse(when, policyFlags);
break;
default:
break;
}
mPointerUsage = POINTER_USAGE_NONE;
}
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
void TouchInputMapper::dispatchPointerGestures(nsecs_t when, uint32_t policyFlags,
bool isTimeout) {
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Update current gesture coordinates.
bool cancelPreviousGesture, finishPreviousGesture;
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
bool sendEvents = preparePointerGestures(when,
&cancelPreviousGesture, &finishPreviousGesture, isTimeout);
if (!sendEvents) {
return;
}
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
if (finishPreviousGesture) {
cancelPreviousGesture = false;
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
// Update the pointer presentation and spots.
if (mParameters.gestureMode == Parameters::GESTURE_MODE_SPOTS) {
mPointerController->setPresentation(PointerControllerInterface::PRESENTATION_SPOT);
if (finishPreviousGesture || cancelPreviousGesture) {
mPointerController->clearSpots();
}
Simplify spot tracking logic. Bug: 4124987 Since spots are now one-to-one with the touch gesture points, we can eliminate a lot of redundant logic that assumed they weren't. Change-Id: Ic60d8d0898c285317f480bc8fb17fb9797770e69
2011-06-06 20:03:18 -07:00
mPointerController->setSpots(mPointerGesture.currentGestureCoords,
mPointerGesture.currentGestureIdToIndex,
mPointerGesture.currentGestureIdBits);
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
} else {
mPointerController->setPresentation(PointerControllerInterface::PRESENTATION_POINTER);
}
Use ViewConfiguration to seed input system configuration. Fix bug where the pointer presentation would be updated on any input reader timeout rather than only when a pointer gesture is in progress. Bug: 4124987 Change-Id: Ie9bba4a0b3228d55e45e65fa2ede5cd6ba887a08
2011-05-26 19:17:02 -07:00
Touch pad UX improvements. Fade the pointer spot when swiping or performing a freeform gesture. Support configuring the gesture mode in the device IDC file. Added workaround for devices that report individual finger movements one at a time instead of all at once. Bug: 4124987 Change-Id: I44628b00382ad59306e4ec5c4717d69cc6fbebb8
2011-05-25 18:23:38 -07:00
// Show or hide the pointer if needed.
switch (mPointerGesture.currentGestureMode) {
case PointerGesture::NEUTRAL:
case PointerGesture::QUIET:
if (mParameters.gestureMode == Parameters::GESTURE_MODE_SPOTS
&& (mPointerGesture.lastGestureMode == PointerGesture::SWIPE
|| mPointerGesture.lastGestureMode == PointerGesture::FREEFORM)) {
// Remind the user of where the pointer is after finishing a gesture with spots.
mPointerController->unfade(PointerControllerInterface::TRANSITION_GRADUAL);
}
break;
case PointerGesture::TAP:
case PointerGesture::TAP_DRAG:
case PointerGesture::BUTTON_CLICK_OR_DRAG:
case PointerGesture::HOVER:
case PointerGesture::PRESS:
// Unfade the pointer when the current gesture manipulates the
// area directly under the pointer.
mPointerController->unfade(PointerControllerInterface::TRANSITION_IMMEDIATE);
break;
case PointerGesture::SWIPE:
case PointerGesture::FREEFORM:
// Fade the pointer when the current gesture manipulates a different
// area and there are spots to guide the user experience.
if (mParameters.gestureMode == Parameters::GESTURE_MODE_SPOTS) {
mPointerController->fade(PointerControllerInterface::TRANSITION_GRADUAL);
} else {
mPointerController->unfade(PointerControllerInterface::TRANSITION_IMMEDIATE);
}
break;
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Send events!
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
int32_t metaState = getContext()->getGlobalMetaState();
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
int32_t buttonState = mCurrentButtonState;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Update last coordinates of pointers that have moved so that we observe the new
// pointer positions at the same time as other pointers that have just gone up.
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
bool down = mPointerGesture.currentGestureMode == PointerGesture::TAP
|| mPointerGesture.currentGestureMode == PointerGesture::TAP_DRAG
|| mPointerGesture.currentGestureMode == PointerGesture::BUTTON_CLICK_OR_DRAG
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
|| mPointerGesture.currentGestureMode == PointerGesture::PRESS
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
|| mPointerGesture.currentGestureMode == PointerGesture::SWIPE
|| mPointerGesture.currentGestureMode == PointerGesture::FREEFORM;
bool moveNeeded = false;
if (down && !cancelPreviousGesture && !finishPreviousGesture
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
&& !mPointerGesture.lastGestureIdBits.isEmpty()
&& !mPointerGesture.currentGestureIdBits.isEmpty()) {
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
BitSet32 movedGestureIdBits(mPointerGesture.currentGestureIdBits.value
& mPointerGesture.lastGestureIdBits.value);
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
moveNeeded = updateMovedPointers(mPointerGesture.currentGestureProperties,
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords, mPointerGesture.currentGestureIdToIndex,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
mPointerGesture.lastGestureProperties,
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.lastGestureCoords, mPointerGesture.lastGestureIdToIndex,
movedGestureIdBits);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (buttonState != mLastButtonState) {
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
moveNeeded = true;
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
// Send motion events for all pointers that went up or were canceled.
BitSet32 dispatchedGestureIdBits(mPointerGesture.lastGestureIdBits);
if (!dispatchedGestureIdBits.isEmpty()) {
if (cancelPreviousGesture) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchMotion(when, policyFlags, mSource,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
AMOTION_EVENT_ACTION_CANCEL, 0, metaState, buttonState,
AMOTION_EVENT_EDGE_FLAG_NONE,
mPointerGesture.lastGestureProperties,
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.lastGestureCoords, mPointerGesture.lastGestureIdToIndex,
dispatchedGestureIdBits, -1,
0, 0, mPointerGesture.downTime);
dispatchedGestureIdBits.clear();
} else {
BitSet32 upGestureIdBits;
if (finishPreviousGesture) {
upGestureIdBits = dispatchedGestureIdBits;
} else {
upGestureIdBits.value = dispatchedGestureIdBits.value
& ~mPointerGesture.currentGestureIdBits.value;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
while (!upGestureIdBits.isEmpty()) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = upGestureIdBits.clearFirstMarkedBit();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchMotion(when, policyFlags, mSource,
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
AMOTION_EVENT_ACTION_POINTER_UP, 0,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
metaState, buttonState, AMOTION_EVENT_EDGE_FLAG_NONE,
mPointerGesture.lastGestureProperties,
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.lastGestureCoords, mPointerGesture.lastGestureIdToIndex,
dispatchedGestureIdBits, id,
0, 0, mPointerGesture.downTime);
dispatchedGestureIdBits.clearBit(id);
}
}
}
// Send motion events for all pointers that moved.
if (moveNeeded) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchMotion(when, policyFlags, mSource,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
AMOTION_EVENT_ACTION_MOVE, 0, metaState, buttonState, AMOTION_EVENT_EDGE_FLAG_NONE,
mPointerGesture.currentGestureProperties,
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords, mPointerGesture.currentGestureIdToIndex,
dispatchedGestureIdBits, -1,
0, 0, mPointerGesture.downTime);
}
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Send motion events for all pointers that went down.
if (down) {
BitSet32 downGestureIdBits(mPointerGesture.currentGestureIdBits.value
& ~dispatchedGestureIdBits.value);
while (!downGestureIdBits.isEmpty()) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = downGestureIdBits.clearFirstMarkedBit();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
dispatchedGestureIdBits.markBit(id);
if (dispatchedGestureIdBits.count() == 1) {
mPointerGesture.downTime = when;
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchMotion(when, policyFlags, mSource,
Remove edge slop support. The edge slop feature was added a long time ago for a specific device that had severe non-linear touch response issues near the edges of its touch screen. The feature is not useful on modern hardware that is property calibrated. Change-Id: I26716cbc7411c31883ddb4bb4ddd981027790d0f
2011-07-14 21:48:23 -07:00
AMOTION_EVENT_ACTION_POINTER_DOWN, 0, metaState, buttonState, 0,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
mPointerGesture.currentGestureProperties,
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords, mPointerGesture.currentGestureIdToIndex,
dispatchedGestureIdBits, id,
0, 0, mPointerGesture.downTime);
}
}
// Send motion events for hover.
if (mPointerGesture.currentGestureMode == PointerGesture::HOVER) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dispatchMotion(when, policyFlags, mSource,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
AMOTION_EVENT_ACTION_HOVER_MOVE, 0,
metaState, buttonState, AMOTION_EVENT_EDGE_FLAG_NONE,
mPointerGesture.currentGestureProperties,
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords, mPointerGesture.currentGestureIdToIndex,
mPointerGesture.currentGestureIdBits, -1,
0, 0, mPointerGesture.downTime);
Improve input event consistency invariants. Fixed some issues where inconsistent streams of events could be generated by the dispatcher, particularly when switching from hovering with one device to hovering with another. Fixed a bug where the touch pad would fail to generate a new HOVER_MOVE following a tap event. As a result, the hover event stream would not resume until the user touched the touch pad again. Change-Id: I444dce84641fb12e56a0af84c931520771d6c493
2011-06-28 20:08:48 -07:00
} else if (dispatchedGestureIdBits.isEmpty()
&& !mPointerGesture.lastGestureIdBits.isEmpty()) {
// Synthesize a hover move event after all pointers go up to indicate that
// the pointer is hovering again even if the user is not currently touching
// the touch pad. This ensures that a view will receive a fresh hover enter
// event after a tap.
float x, y;
mPointerController->getPosition(&x, &y);
PointerProperties pointerProperties;
pointerProperties.clear();
pointerProperties.id = 0;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
pointerProperties.toolType = AMOTION_EVENT_TOOL_TYPE_FINGER;
Improve input event consistency invariants. Fixed some issues where inconsistent streams of events could be generated by the dispatcher, particularly when switching from hovering with one device to hovering with another. Fixed a bug where the touch pad would fail to generate a new HOVER_MOVE following a tap event. As a result, the hover event stream would not resume until the user touched the touch pad again. Change-Id: I444dce84641fb12e56a0af84c931520771d6c493
2011-06-28 20:08:48 -07:00
PointerCoords pointerCoords;
pointerCoords.clear();
pointerCoords.setAxisValue(AMOTION_EVENT_AXIS_X, x);
pointerCoords.setAxisValue(AMOTION_EVENT_AXIS_Y, y);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
NotifyMotionArgs args(when, getDeviceId(), mSource, policyFlags,
Improve input event consistency invariants. Fixed some issues where inconsistent streams of events could be generated by the dispatcher, particularly when switching from hovering with one device to hovering with another. Fixed a bug where the touch pad would fail to generate a new HOVER_MOVE following a tap event. As a result, the hover event stream would not resume until the user touched the touch pad again. Change-Id: I444dce84641fb12e56a0af84c931520771d6c493
2011-06-28 20:08:48 -07:00
AMOTION_EVENT_ACTION_HOVER_MOVE, 0,
metaState, buttonState, AMOTION_EVENT_EDGE_FLAG_NONE,
1, &pointerProperties, &pointerCoords, 0, 0, mPointerGesture.downTime);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
getListener()->notifyMotion(&args);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
// Update state.
mPointerGesture.lastGestureMode = mPointerGesture.currentGestureMode;
if (!down) {
mPointerGesture.lastGestureIdBits.clear();
} else {
mPointerGesture.lastGestureIdBits = mPointerGesture.currentGestureIdBits;
for (BitSet32 idBits(mPointerGesture.currentGestureIdBits); !idBits.isEmpty(); ) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = idBits.clearFirstMarkedBit();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
uint32_t index = mPointerGesture.currentGestureIdToIndex[id];
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
mPointerGesture.lastGestureProperties[index].copyFrom(
mPointerGesture.currentGestureProperties[index]);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.lastGestureCoords[index].copyFrom(
mPointerGesture.currentGestureCoords[index]);
mPointerGesture.lastGestureIdToIndex[id] = index;
}
}
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void TouchInputMapper::abortPointerGestures(nsecs_t when, uint32_t policyFlags) {
// Cancel previously dispatches pointers.
if (!mPointerGesture.lastGestureIdBits.isEmpty()) {
int32_t metaState = getContext()->getGlobalMetaState();
int32_t buttonState = mCurrentButtonState;
dispatchMotion(when, policyFlags, mSource,
AMOTION_EVENT_ACTION_CANCEL, 0, metaState, buttonState,
AMOTION_EVENT_EDGE_FLAG_NONE,
mPointerGesture.lastGestureProperties,
mPointerGesture.lastGestureCoords, mPointerGesture.lastGestureIdToIndex,
mPointerGesture.lastGestureIdBits, -1,
0, 0, mPointerGesture.downTime);
}
// Reset the current pointer gesture.
mPointerGesture.reset();
mPointerVelocityControl.reset();
// Remove any current spots.
if (mPointerController != NULL) {
mPointerController->fade(PointerControllerInterface::TRANSITION_GRADUAL);
mPointerController->clearSpots();
}
}
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
bool TouchInputMapper::preparePointerGestures(nsecs_t when,
bool* outCancelPreviousGesture, bool* outFinishPreviousGesture, bool isTimeout) {
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
*outCancelPreviousGesture = false;
*outFinishPreviousGesture = false;
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
// Handle TAP timeout.
if (isTimeout) {
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: Processing timeout");
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
#endif
if (mPointerGesture.lastGestureMode == PointerGesture::TAP) {
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (when <= mPointerGesture.tapUpTime + mConfig.pointerGestureTapDragInterval) {
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
// The tap/drag timeout has not yet expired.
Use ViewConfiguration to seed input system configuration. Fix bug where the pointer presentation would be updated on any input reader timeout rather than only when a pointer gesture is in progress. Bug: 4124987 Change-Id: Ie9bba4a0b3228d55e45e65fa2ede5cd6ba887a08
2011-05-26 19:17:02 -07:00
getContext()->requestTimeoutAtTime(mPointerGesture.tapUpTime
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
+ mConfig.pointerGestureTapDragInterval);
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
} else {
// The tap is finished.
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: TAP finished");
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
#endif
*outFinishPreviousGesture = true;
mPointerGesture.activeGestureId = -1;
mPointerGesture.currentGestureMode = PointerGesture::NEUTRAL;
mPointerGesture.currentGestureIdBits.clear();
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerVelocityControl.reset();
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
return true;
}
}
// We did not handle this timeout.
return false;
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
const uint32_t currentFingerCount = mCurrentFingerIdBits.count();
const uint32_t lastFingerCount = mLastFingerIdBits.count();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Update the velocity tracker.
{
VelocityTracker::Position positions[MAX_POINTERS];
uint32_t count = 0;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
for (BitSet32 idBits(mCurrentFingerIdBits); !idBits.isEmpty(); count++) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = idBits.clearFirstMarkedBit();
const RawPointerData::Pointer& pointer = mCurrentRawPointerData.pointerForId(id);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
positions[count].x = pointer.x * mPointerXMovementScale;
positions[count].y = pointer.y * mPointerYMovementScale;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mPointerGesture.velocityTracker.addMovement(when,
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mCurrentFingerIdBits, positions);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
// Pick a new active touch id if needed.
// Choose an arbitrary pointer that just went down, if there is one.
// Otherwise choose an arbitrary remaining pointer.
// This guarantees we always have an active touch id when there is at least one pointer.
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
// We keep the same active touch id for as long as possible.
bool activeTouchChanged = false;
int32_t lastActiveTouchId = mPointerGesture.activeTouchId;
int32_t activeTouchId = lastActiveTouchId;
if (activeTouchId < 0) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (!mCurrentFingerIdBits.isEmpty()) {
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
activeTouchChanged = true;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
activeTouchId = mPointerGesture.activeTouchId =
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mCurrentFingerIdBits.firstMarkedBit();
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
mPointerGesture.firstTouchTime = when;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
} else if (!mCurrentFingerIdBits.hasBit(activeTouchId)) {
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
activeTouchChanged = true;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (!mCurrentFingerIdBits.isEmpty()) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
activeTouchId = mPointerGesture.activeTouchId =
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mCurrentFingerIdBits.firstMarkedBit();
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
} else {
activeTouchId = mPointerGesture.activeTouchId = -1;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
}
// Determine whether we are in quiet time.
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
bool isQuietTime = false;
if (activeTouchId < 0) {
mPointerGesture.resetQuietTime();
} else {
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
isQuietTime = when < mPointerGesture.quietTime + mConfig.pointerGestureQuietInterval;
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
if (!isQuietTime) {
if ((mPointerGesture.lastGestureMode == PointerGesture::PRESS
|| mPointerGesture.lastGestureMode == PointerGesture::SWIPE
|| mPointerGesture.lastGestureMode == PointerGesture::FREEFORM)
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
&& currentFingerCount < 2) {
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
// Enter quiet time when exiting swipe or freeform state.
// This is to prevent accidentally entering the hover state and flinging the
// pointer when finishing a swipe and there is still one pointer left onscreen.
isQuietTime = true;
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
} else if (mPointerGesture.lastGestureMode == PointerGesture::BUTTON_CLICK_OR_DRAG
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
&& currentFingerCount >= 2
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
&& !isPointerDown(mCurrentButtonState)) {
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
// Enter quiet time when releasing the button and there are still two or more
// fingers down. This may indicate that one finger was used to press the button
// but it has not gone up yet.
isQuietTime = true;
}
if (isQuietTime) {
mPointerGesture.quietTime = when;
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
}
// Switch states based on button and pointer state.
if (isQuietTime) {
// Case 1: Quiet time. (QUIET)
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: QUIET for next %0.3fms", (mPointerGesture.quietTime
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
+ mConfig.pointerGestureQuietInterval - when) * 0.000001f);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#endif
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
if (mPointerGesture.lastGestureMode != PointerGesture::QUIET) {
*outFinishPreviousGesture = true;
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.activeGestureId = -1;
mPointerGesture.currentGestureMode = PointerGesture::QUIET;
mPointerGesture.currentGestureIdBits.clear();
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerVelocityControl.reset();
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
} else if (isPointerDown(mCurrentButtonState)) {
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
// Case 2: Button is pressed. (BUTTON_CLICK_OR_DRAG)
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// The pointer follows the active touch point.
// Emit DOWN, MOVE, UP events at the pointer location.
//
// Only the active touch matters; other fingers are ignored. This policy helps
// to handle the case where the user places a second finger on the touch pad
// to apply the necessary force to depress an integrated button below the surface.
// We don't want the second finger to be delivered to applications.
//
// For this to work well, we need to make sure to track the pointer that is really
// active. If the user first puts one finger down to click then adds another
// finger to drag then the active pointer should switch to the finger that is
// being dragged.
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: BUTTON_CLICK_OR_DRAG activeTouchId=%d, "
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
"currentFingerCount=%d", activeTouchId, currentFingerCount);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#endif
// Reset state when just starting.
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
if (mPointerGesture.lastGestureMode != PointerGesture::BUTTON_CLICK_OR_DRAG) {
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
*outFinishPreviousGesture = true;
mPointerGesture.activeGestureId = 0;
}
// Switch pointers if needed.
// Find the fastest pointer and follow it.
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (activeTouchId >= 0 && currentFingerCount > 1) {
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
int32_t bestId = -1;
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
float bestSpeed = mConfig.pointerGestureDragMinSwitchSpeed;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
for (BitSet32 idBits(mCurrentFingerIdBits); !idBits.isEmpty(); ) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = idBits.clearFirstMarkedBit();
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
float vx, vy;
if (mPointerGesture.velocityTracker.getVelocity(id, &vx, &vy)) {
float speed = hypotf(vx, vy);
if (speed > bestSpeed) {
bestId = id;
bestSpeed = speed;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
}
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
}
if (bestId >= 0 && bestId != activeTouchId) {
mPointerGesture.activeTouchId = activeTouchId = bestId;
activeTouchChanged = true;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: BUTTON_CLICK_OR_DRAG switched pointers, "
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
"bestId=%d, bestSpeed=%0.3f", bestId, bestSpeed);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#endif
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
}
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
}
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (activeTouchId >= 0 && mLastFingerIdBits.hasBit(activeTouchId)) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
const RawPointerData::Pointer& currentPointer =
mCurrentRawPointerData.pointerForId(activeTouchId);
const RawPointerData::Pointer& lastPointer =
mLastRawPointerData.pointerForId(activeTouchId);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
float deltaX = (currentPointer.x - lastPointer.x) * mPointerXMovementScale;
float deltaY = (currentPointer.y - lastPointer.y) * mPointerYMovementScale;
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
rotateDelta(mSurfaceOrientation, &deltaX, &deltaY);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerVelocityControl.move(when, &deltaX, &deltaY);
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
// Move the pointer using a relative motion.
// When using spots, the click will occur at the position of the anchor
// spot and all other spots will move there.
mPointerController->move(deltaX, deltaY);
} else {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerVelocityControl.reset();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
float x, y;
mPointerController->getPosition(&x, &y);
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
mPointerGesture.currentGestureMode = PointerGesture::BUTTON_CLICK_OR_DRAG;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureIdBits.clear();
mPointerGesture.currentGestureIdBits.markBit(mPointerGesture.activeGestureId);
mPointerGesture.currentGestureIdToIndex[mPointerGesture.activeGestureId] = 0;
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
mPointerGesture.currentGestureProperties[0].clear();
mPointerGesture.currentGestureProperties[0].id = mPointerGesture.activeGestureId;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
mPointerGesture.currentGestureProperties[0].toolType = AMOTION_EVENT_TOOL_TYPE_FINGER;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords[0].clear();
mPointerGesture.currentGestureCoords[0].setAxisValue(AMOTION_EVENT_AXIS_X, x);
mPointerGesture.currentGestureCoords[0].setAxisValue(AMOTION_EVENT_AXIS_Y, y);
mPointerGesture.currentGestureCoords[0].setAxisValue(AMOTION_EVENT_AXIS_PRESSURE, 1.0f);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
} else if (currentFingerCount == 0) {
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// Case 3. No fingers down and button is not pressed. (NEUTRAL)
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
if (mPointerGesture.lastGestureMode != PointerGesture::NEUTRAL) {
*outFinishPreviousGesture = true;
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
// Watch for taps coming out of HOVER or TAP_DRAG mode.
Use ViewConfiguration to seed input system configuration. Fix bug where the pointer presentation would be updated on any input reader timeout rather than only when a pointer gesture is in progress. Bug: 4124987 Change-Id: Ie9bba4a0b3228d55e45e65fa2ede5cd6ba887a08
2011-05-26 19:17:02 -07:00
// Checking for taps after TAP_DRAG allows us to detect double-taps.
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
bool tapped = false;
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
if ((mPointerGesture.lastGestureMode == PointerGesture::HOVER
|| mPointerGesture.lastGestureMode == PointerGesture::TAP_DRAG)
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
&& lastFingerCount == 1) {
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (when <= mPointerGesture.tapDownTime + mConfig.pointerGestureTapInterval) {
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
float x, y;
mPointerController->getPosition(&x, &y);
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (fabs(x - mPointerGesture.tapX) <= mConfig.pointerGestureTapSlop
&& fabs(y - mPointerGesture.tapY) <= mConfig.pointerGestureTapSlop) {
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: TAP");
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#endif
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
mPointerGesture.tapUpTime = when;
Use ViewConfiguration to seed input system configuration. Fix bug where the pointer presentation would be updated on any input reader timeout rather than only when a pointer gesture is in progress. Bug: 4124987 Change-Id: Ie9bba4a0b3228d55e45e65fa2ede5cd6ba887a08
2011-05-26 19:17:02 -07:00
getContext()->requestTimeoutAtTime(when
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
+ mConfig.pointerGestureTapDragInterval);
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.activeGestureId = 0;
mPointerGesture.currentGestureMode = PointerGesture::TAP;
mPointerGesture.currentGestureIdBits.clear();
mPointerGesture.currentGestureIdBits.markBit(
mPointerGesture.activeGestureId);
mPointerGesture.currentGestureIdToIndex[
mPointerGesture.activeGestureId] = 0;
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
mPointerGesture.currentGestureProperties[0].clear();
mPointerGesture.currentGestureProperties[0].id =
mPointerGesture.activeGestureId;
mPointerGesture.currentGestureProperties[0].toolType =
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
AMOTION_EVENT_TOOL_TYPE_FINGER;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords[0].clear();
mPointerGesture.currentGestureCoords[0].setAxisValue(
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
AMOTION_EVENT_AXIS_X, mPointerGesture.tapX);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords[0].setAxisValue(
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
AMOTION_EVENT_AXIS_Y, mPointerGesture.tapY);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords[0].setAxisValue(
AMOTION_EVENT_AXIS_PRESSURE, 1.0f);
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
tapped = true;
Add support for detecting finger orientation. Change-Id: I901fc61f7705fcd7950190dc8782bad616032ee2
2011-01-14 19:09:23 -08:00
} else {
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: Not a TAP, deltaX=%f, deltaY=%f",
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
x - mPointerGesture.tapX,
y - mPointerGesture.tapY);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#endif
Add support for detecting finger orientation. Change-Id: I901fc61f7705fcd7950190dc8782bad616032ee2
2011-01-14 19:09:23 -08:00
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
} else {
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: Not a TAP, %0.3fms since down",
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
(when - mPointerGesture.tapDownTime) * 0.000001f);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#endif
Add support for detecting finger orientation. Change-Id: I901fc61f7705fcd7950190dc8782bad616032ee2
2011-01-14 19:09:23 -08:00
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerVelocityControl.reset();
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
if (!tapped) {
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: NEUTRAL");
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#endif
mPointerGesture.activeGestureId = -1;
mPointerGesture.currentGestureMode = PointerGesture::NEUTRAL;
mPointerGesture.currentGestureIdBits.clear();
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
} else if (currentFingerCount == 1) {
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
// Case 4. Exactly one finger down, button is not pressed. (HOVER or TAP_DRAG)
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
// The pointer follows the active touch point.
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
// When in HOVER, emit HOVER_MOVE events at the pointer location.
// When in TAP_DRAG, emit MOVE events at the pointer location.
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(activeTouchId >= 0);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
mPointerGesture.currentGestureMode = PointerGesture::HOVER;
if (mPointerGesture.lastGestureMode == PointerGesture::TAP) {
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (when <= mPointerGesture.tapUpTime + mConfig.pointerGestureTapDragInterval) {
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
float x, y;
mPointerController->getPosition(&x, &y);
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (fabs(x - mPointerGesture.tapX) <= mConfig.pointerGestureTapSlop
&& fabs(y - mPointerGesture.tapY) <= mConfig.pointerGestureTapSlop) {
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
mPointerGesture.currentGestureMode = PointerGesture::TAP_DRAG;
} else {
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: Not a TAP_DRAG, deltaX=%f, deltaY=%f",
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
x - mPointerGesture.tapX,
y - mPointerGesture.tapY);
#endif
}
} else {
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: Not a TAP_DRAG, %0.3fms time since up",
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
(when - mPointerGesture.tapUpTime) * 0.000001f);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#endif
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
}
} else if (mPointerGesture.lastGestureMode == PointerGesture::TAP_DRAG) {
mPointerGesture.currentGestureMode = PointerGesture::TAP_DRAG;
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (mLastFingerIdBits.hasBit(activeTouchId)) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
const RawPointerData::Pointer& currentPointer =
mCurrentRawPointerData.pointerForId(activeTouchId);
const RawPointerData::Pointer& lastPointer =
mLastRawPointerData.pointerForId(activeTouchId);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
float deltaX = (currentPointer.x - lastPointer.x)
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
* mPointerXMovementScale;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
float deltaY = (currentPointer.y - lastPointer.y)
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
* mPointerYMovementScale;
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
rotateDelta(mSurfaceOrientation, &deltaX, &deltaY);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerVelocityControl.move(when, &deltaX, &deltaY);
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
// Move the pointer using a relative motion.
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
// When using spots, the hover or drag will occur at the position of the anchor spot.
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerController->move(deltaX, deltaY);
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
} else {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerVelocityControl.reset();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
bool down;
if (mPointerGesture.currentGestureMode == PointerGesture::TAP_DRAG) {
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: TAP_DRAG");
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
#endif
down = true;
} else {
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: HOVER");
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
#endif
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
if (mPointerGesture.lastGestureMode != PointerGesture::HOVER) {
*outFinishPreviousGesture = true;
}
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
mPointerGesture.activeGestureId = 0;
down = false;
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
float x, y;
mPointerController->getPosition(&x, &y);
mPointerGesture.currentGestureIdBits.clear();
mPointerGesture.currentGestureIdBits.markBit(mPointerGesture.activeGestureId);
mPointerGesture.currentGestureIdToIndex[mPointerGesture.activeGestureId] = 0;
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
mPointerGesture.currentGestureProperties[0].clear();
mPointerGesture.currentGestureProperties[0].id = mPointerGesture.activeGestureId;
mPointerGesture.currentGestureProperties[0].toolType =
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
AMOTION_EVENT_TOOL_TYPE_FINGER;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords[0].clear();
mPointerGesture.currentGestureCoords[0].setAxisValue(AMOTION_EVENT_AXIS_X, x);
mPointerGesture.currentGestureCoords[0].setAxisValue(AMOTION_EVENT_AXIS_Y, y);
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
mPointerGesture.currentGestureCoords[0].setAxisValue(AMOTION_EVENT_AXIS_PRESSURE,
down ? 1.0f : 0.0f);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (lastFingerCount == 0 && currentFingerCount != 0) {
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
mPointerGesture.resetTap();
mPointerGesture.tapDownTime = when;
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
mPointerGesture.tapX = x;
mPointerGesture.tapY = y;
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
} else {
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
// Case 5. At least two fingers down, button is not pressed. (PRESS, SWIPE or FREEFORM)
// We need to provide feedback for each finger that goes down so we cannot wait
// for the fingers to move before deciding what to do.
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
//
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
// The ambiguous case is deciding what to do when there are two fingers down but they
// have not moved enough to determine whether they are part of a drag or part of a
// freeform gesture, or just a press or long-press at the pointer location.
//
// When there are two fingers we start with the PRESS hypothesis and we generate a
// down at the pointer location.
//
// When the two fingers move enough or when additional fingers are added, we make
// a decision to transition into SWIPE or FREEFORM mode accordingly.
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(activeTouchId >= 0);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Use ViewConfiguration to seed input system configuration. Fix bug where the pointer presentation would be updated on any input reader timeout rather than only when a pointer gesture is in progress. Bug: 4124987 Change-Id: Ie9bba4a0b3228d55e45e65fa2ede5cd6ba887a08
2011-05-26 19:17:02 -07:00
bool settled = when >= mPointerGesture.firstTouchTime
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
+ mConfig.pointerGestureMultitouchSettleInterval;
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
if (mPointerGesture.lastGestureMode != PointerGesture::PRESS
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
&& mPointerGesture.lastGestureMode != PointerGesture::SWIPE
&& mPointerGesture.lastGestureMode != PointerGesture::FREEFORM) {
*outFinishPreviousGesture = true;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
} else if (!settled && currentFingerCount > lastFingerCount) {
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
// Additional pointers have gone down but not yet settled.
// Reset the gesture.
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: Resetting gesture since additional pointers went down for MULTITOUCH, "
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
"settle time remaining %0.3fms", (mPointerGesture.firstTouchTime
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
+ mConfig.pointerGestureMultitouchSettleInterval - when)
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
* 0.000001f);
#endif
*outCancelPreviousGesture = true;
} else {
// Continue previous gesture.
mPointerGesture.currentGestureMode = mPointerGesture.lastGestureMode;
}
if (*outFinishPreviousGesture || *outCancelPreviousGesture) {
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
mPointerGesture.currentGestureMode = PointerGesture::PRESS;
mPointerGesture.activeGestureId = 0;
Touch pad UX improvements. Fade the pointer spot when swiping or performing a freeform gesture. Support configuring the gesture mode in the device IDC file. Added workaround for devices that report individual finger movements one at a time instead of all at once. Bug: 4124987 Change-Id: I44628b00382ad59306e4ec5c4717d69cc6fbebb8
2011-05-25 18:23:38 -07:00
mPointerGesture.referenceIdBits.clear();
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerVelocityControl.reset();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Simplify spot tracking logic. Bug: 4124987 Since spots are now one-to-one with the touch gesture points, we can eliminate a lot of redundant logic that assumed they weren't. Change-Id: Ic60d8d0898c285317f480bc8fb17fb9797770e69
2011-06-06 20:03:18 -07:00
// Use the centroid and pointer location as the reference points for the gesture.
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: Using centroid as reference for MULTITOUCH, "
Simplify spot tracking logic. Bug: 4124987 Since spots are now one-to-one with the touch gesture points, we can eliminate a lot of redundant logic that assumed they weren't. Change-Id: Ic60d8d0898c285317f480bc8fb17fb9797770e69
2011-06-06 20:03:18 -07:00
"settle time remaining %0.3fms", (mPointerGesture.firstTouchTime
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
+ mConfig.pointerGestureMultitouchSettleInterval - when)
Simplify spot tracking logic. Bug: 4124987 Since spots are now one-to-one with the touch gesture points, we can eliminate a lot of redundant logic that assumed they weren't. Change-Id: Ic60d8d0898c285317f480bc8fb17fb9797770e69
2011-06-06 20:03:18 -07:00
* 0.000001f);
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
#endif
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentRawPointerData.getCentroidOfTouchingPointers(
&mPointerGesture.referenceTouchX,
Simplify spot tracking logic. Bug: 4124987 Since spots are now one-to-one with the touch gesture points, we can eliminate a lot of redundant logic that assumed they weren't. Change-Id: Ic60d8d0898c285317f480bc8fb17fb9797770e69
2011-06-06 20:03:18 -07:00
&mPointerGesture.referenceTouchY);
mPointerController->getPosition(&mPointerGesture.referenceGestureX,
&mPointerGesture.referenceGestureY);
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
// Clear the reference deltas for fingers not yet included in the reference calculation.
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
for (BitSet32 idBits(mCurrentFingerIdBits.value
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
& ~mPointerGesture.referenceIdBits.value); !idBits.isEmpty(); ) {
uint32_t id = idBits.clearFirstMarkedBit();
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
mPointerGesture.referenceDeltas[id].dx = 0;
mPointerGesture.referenceDeltas[id].dy = 0;
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerGesture.referenceIdBits = mCurrentFingerIdBits;
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
// Add delta for all fingers and calculate a common movement delta.
float commonDeltaX = 0, commonDeltaY = 0;
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
BitSet32 commonIdBits(mLastFingerIdBits.value
& mCurrentFingerIdBits.value);
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
for (BitSet32 idBits(commonIdBits); !idBits.isEmpty(); ) {
bool first = (idBits == commonIdBits);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = idBits.clearFirstMarkedBit();
const RawPointerData::Pointer& cpd = mCurrentRawPointerData.pointerForId(id);
const RawPointerData::Pointer& lpd = mLastRawPointerData.pointerForId(id);
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
PointerGesture::Delta& delta = mPointerGesture.referenceDeltas[id];
delta.dx += cpd.x - lpd.x;
delta.dy += cpd.y - lpd.y;
if (first) {
commonDeltaX = delta.dx;
commonDeltaY = delta.dy;
} else {
commonDeltaX = calculateCommonVector(commonDeltaX, delta.dx);
commonDeltaY = calculateCommonVector(commonDeltaY, delta.dy);
}
}
// Consider transitions from PRESS to SWIPE or MULTITOUCH.
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
if (mPointerGesture.currentGestureMode == PointerGesture::PRESS) {
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
float dist[MAX_POINTER_ID + 1];
int32_t distOverThreshold = 0;
for (BitSet32 idBits(mPointerGesture.referenceIdBits); !idBits.isEmpty(); ) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = idBits.clearFirstMarkedBit();
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
PointerGesture::Delta& delta = mPointerGesture.referenceDeltas[id];
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
dist[id] = hypotf(delta.dx * mPointerXZoomScale,
delta.dy * mPointerYZoomScale);
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (dist[id] > mConfig.pointerGestureMultitouchMinDistance) {
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
distOverThreshold += 1;
}
}
// Only transition when at least two pointers have moved further than
// the minimum distance threshold.
if (distOverThreshold >= 2) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (currentFingerCount > 2) {
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
// There are more than two pointers, switch to FREEFORM.
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: PRESS transitioned to FREEFORM, number of pointers %d > 2",
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
currentFingerCount);
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
#endif
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
*outCancelPreviousGesture = true;
mPointerGesture.currentGestureMode = PointerGesture::FREEFORM;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
} else {
// There are exactly two pointers.
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
BitSet32 idBits(mCurrentFingerIdBits);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id1 = idBits.clearFirstMarkedBit();
uint32_t id2 = idBits.firstMarkedBit();
const RawPointerData::Pointer& p1 = mCurrentRawPointerData.pointerForId(id1);
const RawPointerData::Pointer& p2 = mCurrentRawPointerData.pointerForId(id2);
float mutualDistance = distance(p1.x, p1.y, p2.x, p2.y);
if (mutualDistance > mPointerGestureMaxSwipeWidth) {
// There are two pointers but they are too far apart for a SWIPE,
// switch to FREEFORM.
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: PRESS transitioned to FREEFORM, distance %0.3f > %0.3f",
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mutualDistance, mPointerGestureMaxSwipeWidth);
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
#endif
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
*outCancelPreviousGesture = true;
mPointerGesture.currentGestureMode = PointerGesture::FREEFORM;
} else {
// There are two pointers. Wait for both pointers to start moving
// before deciding whether this is a SWIPE or FREEFORM gesture.
float dist1 = dist[id1];
float dist2 = dist[id2];
if (dist1 >= mConfig.pointerGestureMultitouchMinDistance
&& dist2 >= mConfig.pointerGestureMultitouchMinDistance) {
// Calculate the dot product of the displacement vectors.
// When the vectors are oriented in approximately the same direction,
// the angle betweeen them is near zero and the cosine of the angle
// approches 1.0. Recall that dot(v1, v2) = cos(angle) * mag(v1) * mag(v2).
PointerGesture::Delta& delta1 = mPointerGesture.referenceDeltas[id1];
PointerGesture::Delta& delta2 = mPointerGesture.referenceDeltas[id2];
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
float dx1 = delta1.dx * mPointerXZoomScale;
float dy1 = delta1.dy * mPointerYZoomScale;
float dx2 = delta2.dx * mPointerXZoomScale;
float dy2 = delta2.dy * mPointerYZoomScale;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
float dot = dx1 * dx2 + dy1 * dy2;
float cosine = dot / (dist1 * dist2); // denominator always > 0
if (cosine >= mConfig.pointerGestureSwipeTransitionAngleCosine) {
// Pointers are moving in the same direction. Switch to SWIPE.
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: PRESS transitioned to SWIPE, "
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
"dist1 %0.3f >= %0.3f, dist2 %0.3f >= %0.3f, "
"cosine %0.3f >= %0.3f",
dist1, mConfig.pointerGestureMultitouchMinDistance,
dist2, mConfig.pointerGestureMultitouchMinDistance,
cosine, mConfig.pointerGestureSwipeTransitionAngleCosine);
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
#endif
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mPointerGesture.currentGestureMode = PointerGesture::SWIPE;
} else {
// Pointers are moving in different directions. Switch to FREEFORM.
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: PRESS transitioned to FREEFORM, "
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
"dist1 %0.3f >= %0.3f, dist2 %0.3f >= %0.3f, "
"cosine %0.3f < %0.3f",
dist1, mConfig.pointerGestureMultitouchMinDistance,
dist2, mConfig.pointerGestureMultitouchMinDistance,
cosine, mConfig.pointerGestureSwipeTransitionAngleCosine);
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
#endif
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
*outCancelPreviousGesture = true;
mPointerGesture.currentGestureMode = PointerGesture::FREEFORM;
}
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
}
}
} else if (mPointerGesture.currentGestureMode == PointerGesture::SWIPE) {
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
// Switch from SWIPE to FREEFORM if additional pointers go down.
// Cancel previous gesture.
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (currentFingerCount > 2) {
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: SWIPE transitioned to FREEFORM, number of pointers %d > 2",
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
currentFingerCount);
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
#endif
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
*outCancelPreviousGesture = true;
mPointerGesture.currentGestureMode = PointerGesture::FREEFORM;
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
// Move the reference points based on the overall group motion of the fingers
// except in PRESS mode while waiting for a transition to occur.
if (mPointerGesture.currentGestureMode != PointerGesture::PRESS
&& (commonDeltaX || commonDeltaY)) {
for (BitSet32 idBits(mPointerGesture.referenceIdBits); !idBits.isEmpty(); ) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = idBits.clearFirstMarkedBit();
Touch pad UX improvements. Fade the pointer spot when swiping or performing a freeform gesture. Support configuring the gesture mode in the device IDC file. Added workaround for devices that report individual finger movements one at a time instead of all at once. Bug: 4124987 Change-Id: I44628b00382ad59306e4ec5c4717d69cc6fbebb8
2011-05-25 18:23:38 -07:00
PointerGesture::Delta& delta = mPointerGesture.referenceDeltas[id];
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
delta.dx = 0;
delta.dy = 0;
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
}
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
mPointerGesture.referenceTouchX += commonDeltaX;
mPointerGesture.referenceTouchY += commonDeltaY;
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
commonDeltaX *= mPointerXMovementScale;
commonDeltaY *= mPointerYMovementScale;
Add support for orientationAware touch pads. Bug: 5038388 Change-Id: I9c571b1542299334d2d64c5b33e456ec20ebc0ae
2011-07-15 20:44:17 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
rotateDelta(mSurfaceOrientation, &commonDeltaX, &commonDeltaY);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerVelocityControl.move(when, &commonDeltaX, &commonDeltaY);
Implement pointer acceleration. Bug: 4124987 Change-Id: I1f31a28f1594c55302ccabe13fe3ca6d2ff71d50
2011-06-01 12:33:19 -07:00
Touch pad improvements. Bug: 4124987 Only show one spot per touch point instead of one spot per finger for multitouch gestures. Tweaked the pointer acceleration curves. Dissociated the hover/tap timeouts from the "tap" timeout since they mean very different things. Change-Id: I7c2cbd30feeb65ebc12f6c7e33a67dc9a9f59d4c
2011-06-06 19:23:05 -07:00
mPointerGesture.referenceGestureX += commonDeltaX;
mPointerGesture.referenceGestureY += commonDeltaY;
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
}
// Report gestures.
Add support for orientationAware touch pads. Bug: 5038388 Change-Id: I9c571b1542299334d2d64c5b33e456ec20ebc0ae
2011-07-15 20:44:17 -07:00
if (mPointerGesture.currentGestureMode == PointerGesture::PRESS
|| mPointerGesture.currentGestureMode == PointerGesture::SWIPE) {
// PRESS or SWIPE mode.
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: PRESS or SWIPE activeTouchId=%d,"
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
"activeGestureId=%d, currentTouchPointerCount=%d",
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
activeTouchId, mPointerGesture.activeGestureId, currentFingerCount);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#endif
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(mPointerGesture.activeGestureId >= 0);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureIdBits.clear();
mPointerGesture.currentGestureIdBits.markBit(mPointerGesture.activeGestureId);
mPointerGesture.currentGestureIdToIndex[mPointerGesture.activeGestureId] = 0;
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
mPointerGesture.currentGestureProperties[0].clear();
mPointerGesture.currentGestureProperties[0].id = mPointerGesture.activeGestureId;
mPointerGesture.currentGestureProperties[0].toolType =
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
AMOTION_EVENT_TOOL_TYPE_FINGER;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords[0].clear();
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
mPointerGesture.currentGestureCoords[0].setAxisValue(AMOTION_EVENT_AXIS_X,
mPointerGesture.referenceGestureX);
mPointerGesture.currentGestureCoords[0].setAxisValue(AMOTION_EVENT_AXIS_Y,
mPointerGesture.referenceGestureY);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords[0].setAxisValue(AMOTION_EVENT_AXIS_PRESSURE, 1.0f);
} else if (mPointerGesture.currentGestureMode == PointerGesture::FREEFORM) {
// FREEFORM mode.
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: FREEFORM activeTouchId=%d,"
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
"activeGestureId=%d, currentTouchPointerCount=%d",
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
activeTouchId, mPointerGesture.activeGestureId, currentFingerCount);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#endif
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(mPointerGesture.activeGestureId >= 0);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureIdBits.clear();
BitSet32 mappedTouchIdBits;
BitSet32 usedGestureIdBits;
if (mPointerGesture.lastGestureMode != PointerGesture::FREEFORM) {
// Initially, assign the active gesture id to the active touch point
// if there is one. No other touch id bits are mapped yet.
if (!*outCancelPreviousGesture) {
mappedTouchIdBits.markBit(activeTouchId);
usedGestureIdBits.markBit(mPointerGesture.activeGestureId);
mPointerGesture.freeformTouchToGestureIdMap[activeTouchId] =
mPointerGesture.activeGestureId;
} else {
mPointerGesture.activeGestureId = -1;
}
} else {
// Otherwise, assume we mapped all touches from the previous frame.
// Reuse all mappings that are still applicable.
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mappedTouchIdBits.value = mLastFingerIdBits.value
& mCurrentFingerIdBits.value;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
usedGestureIdBits = mPointerGesture.lastGestureIdBits;
// Check whether we need to choose a new active gesture id because the
// current went went up.
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
for (BitSet32 upTouchIdBits(mLastFingerIdBits.value
& ~mCurrentFingerIdBits.value);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
!upTouchIdBits.isEmpty(); ) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t upTouchId = upTouchIdBits.clearFirstMarkedBit();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
uint32_t upGestureId = mPointerGesture.freeformTouchToGestureIdMap[upTouchId];
if (upGestureId == uint32_t(mPointerGesture.activeGestureId)) {
mPointerGesture.activeGestureId = -1;
break;
}
}
}
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: FREEFORM follow up "
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
"mappedTouchIdBits=0x%08x, usedGestureIdBits=0x%08x, "
"activeGestureId=%d",
mappedTouchIdBits.value, usedGestureIdBits.value,
mPointerGesture.activeGestureId);
#endif
Add support for arbitrary axes in MotionEvents. This change makes it possible to extend the set of axes that are reported in MotionEvents by defining new axis constants. The MotionEvent object is now backed by its C++ counterpart to avoid having to maintain multiple representations of the same data. Change-Id: Ibe93c90d4b390d43c176cce48d558d20869ee608
2011-02-14 17:03:18 -08:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
BitSet32 idBits(mCurrentFingerIdBits);
for (uint32_t i = 0; i < currentFingerCount; i++) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t touchId = idBits.clearFirstMarkedBit();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
uint32_t gestureId;
if (!mappedTouchIdBits.hasBit(touchId)) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
gestureId = usedGestureIdBits.markFirstUnmarkedBit();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.freeformTouchToGestureIdMap[touchId] = gestureId;
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: FREEFORM "
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
"new mapping for touch id %d -> gesture id %d",
touchId, gestureId);
#endif
} else {
gestureId = mPointerGesture.freeformTouchToGestureIdMap[touchId];
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: FREEFORM "
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
"existing mapping for touch id %d -> gesture id %d",
touchId, gestureId);
#endif
}
mPointerGesture.currentGestureIdBits.markBit(gestureId);
mPointerGesture.currentGestureIdToIndex[gestureId] = i;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
const RawPointerData::Pointer& pointer =
mCurrentRawPointerData.pointerForId(touchId);
float deltaX = (pointer.x - mPointerGesture.referenceTouchX)
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
* mPointerXZoomScale;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
float deltaY = (pointer.y - mPointerGesture.referenceTouchY)
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
* mPointerYZoomScale;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
rotateDelta(mSurfaceOrientation, &deltaX, &deltaY);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
mPointerGesture.currentGestureProperties[i].clear();
mPointerGesture.currentGestureProperties[i].id = gestureId;
mPointerGesture.currentGestureProperties[i].toolType =
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
AMOTION_EVENT_TOOL_TYPE_FINGER;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords[i].clear();
mPointerGesture.currentGestureCoords[i].setAxisValue(
Add support for orientationAware touch pads. Bug: 5038388 Change-Id: I9c571b1542299334d2d64c5b33e456ec20ebc0ae
2011-07-15 20:44:17 -07:00
AMOTION_EVENT_AXIS_X, mPointerGesture.referenceGestureX + deltaX);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords[i].setAxisValue(
Add support for orientationAware touch pads. Bug: 5038388 Change-Id: I9c571b1542299334d2d64c5b33e456ec20ebc0ae
2011-07-15 20:44:17 -07:00
AMOTION_EVENT_AXIS_Y, mPointerGesture.referenceGestureY + deltaY);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
mPointerGesture.currentGestureCoords[i].setAxisValue(
AMOTION_EVENT_AXIS_PRESSURE, 1.0f);
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
if (mPointerGesture.activeGestureId < 0) {
mPointerGesture.activeGestureId =
mPointerGesture.currentGestureIdBits.firstMarkedBit();
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: FREEFORM new "
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
"activeGestureId=%d", mPointerGesture.activeGestureId);
#endif
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
}
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mPointerController->setButtonState(mCurrentButtonState);
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
#if DEBUG_GESTURES
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("Gestures: finishPreviousGesture=%s, cancelPreviousGesture=%s, "
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
"currentGestureMode=%d, currentGestureIdBits=0x%08x, "
"lastGestureMode=%d, lastGestureIdBits=0x%08x",
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
toString(*outFinishPreviousGesture), toString(*outCancelPreviousGesture),
Initial checkin of spot presentation for touchpad gestures. Added a new PointerIcon API (hidden for now) for loading pointer icons. Fixed a starvation problem in the native Looper's sendMessage implementation which caused new messages to be posted ahead of old messages sent with sendMessageDelayed. Redesigned the touch pad gestures to be defined in terms of more fluid finger / spot movements. The objective is to reinforce the natural mapping between fingers and spots which means there must not be any discontinuities in spot motion relative to the fingers. Removed the SpotController stub and folded its responsibilities into PointerController. Change-Id: I5126b1e69d95252fda7f2a684c9287e239a57163
2011-04-12 22:39:53 -07:00
mPointerGesture.currentGestureMode, mPointerGesture.currentGestureIdBits.value,
mPointerGesture.lastGestureMode, mPointerGesture.lastGestureIdBits.value);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
for (BitSet32 idBits = mPointerGesture.currentGestureIdBits; !idBits.isEmpty(); ) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = idBits.clearFirstMarkedBit();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
uint32_t index = mPointerGesture.currentGestureIdToIndex[id];
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
const PointerProperties& properties = mPointerGesture.currentGestureProperties[index];
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
const PointerCoords& coords = mPointerGesture.currentGestureCoords[index];
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD(" currentGesture[%d]: index=%d, toolType=%d, "
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
"x=%0.3f, y=%0.3f, pressure=%0.3f",
id, index, properties.toolType,
coords.getAxisValue(AMOTION_EVENT_AXIS_X),
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
coords.getAxisValue(AMOTION_EVENT_AXIS_Y),
coords.getAxisValue(AMOTION_EVENT_AXIS_PRESSURE));
}
for (BitSet32 idBits = mPointerGesture.lastGestureIdBits; !idBits.isEmpty(); ) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = idBits.clearFirstMarkedBit();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
uint32_t index = mPointerGesture.lastGestureIdToIndex[id];
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
const PointerProperties& properties = mPointerGesture.lastGestureProperties[index];
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
const PointerCoords& coords = mPointerGesture.lastGestureCoords[index];
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD(" lastGesture[%d]: index=%d, toolType=%d, "
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
"x=%0.3f, y=%0.3f, pressure=%0.3f",
id, index, properties.toolType,
coords.getAxisValue(AMOTION_EVENT_AXIS_X),
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
coords.getAxisValue(AMOTION_EVENT_AXIS_Y),
coords.getAxisValue(AMOTION_EVENT_AXIS_PRESSURE));
}
#endif
Add tap/drag touchpad gesture. The mapper sends a DOWN when the finger is released then starts a short timer. If the finger is not pressed again before the timer elapses, then the mapper sends an UP to complete the tap. If the finger is pressed again then the mapper starts sending MOVEs as part of a drag. Double/triple taps work as intended because we also start watching for a new tap when the finger is pressed again. If a new tap occurs the old tap is also finished. So each individual finger short press/release cycle constitutes a distinct tap. Change-Id: Iaa51182edee4378162a66adfd746088e93af2a30
2011-04-19 21:20:10 -07:00
return true;
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void TouchInputMapper::dispatchPointerStylus(nsecs_t when, uint32_t policyFlags) {
mPointerSimple.currentCoords.clear();
mPointerSimple.currentProperties.clear();
bool down, hovering;
if (!mCurrentStylusIdBits.isEmpty()) {
uint32_t id = mCurrentStylusIdBits.firstMarkedBit();
uint32_t index = mCurrentCookedPointerData.idToIndex[id];
float x = mCurrentCookedPointerData.pointerCoords[index].getX();
float y = mCurrentCookedPointerData.pointerCoords[index].getY();
mPointerController->setPosition(x, y);
hovering = mCurrentCookedPointerData.hoveringIdBits.hasBit(id);
down = !hovering;
mPointerController->getPosition(&x, &y);
mPointerSimple.currentCoords.copyFrom(mCurrentCookedPointerData.pointerCoords[index]);
mPointerSimple.currentCoords.setAxisValue(AMOTION_EVENT_AXIS_X, x);
mPointerSimple.currentCoords.setAxisValue(AMOTION_EVENT_AXIS_Y, y);
mPointerSimple.currentProperties.id = 0;
mPointerSimple.currentProperties.toolType =
mCurrentCookedPointerData.pointerProperties[index].toolType;
} else {
down = false;
hovering = false;
}
dispatchPointerSimple(when, policyFlags, down, hovering);
}
void TouchInputMapper::abortPointerStylus(nsecs_t when, uint32_t policyFlags) {
abortPointerSimple(when, policyFlags);
}
void TouchInputMapper::dispatchPointerMouse(nsecs_t when, uint32_t policyFlags) {
mPointerSimple.currentCoords.clear();
mPointerSimple.currentProperties.clear();
bool down, hovering;
if (!mCurrentMouseIdBits.isEmpty()) {
uint32_t id = mCurrentMouseIdBits.firstMarkedBit();
uint32_t currentIndex = mCurrentRawPointerData.idToIndex[id];
if (mLastMouseIdBits.hasBit(id)) {
uint32_t lastIndex = mCurrentRawPointerData.idToIndex[id];
float deltaX = (mCurrentRawPointerData.pointers[currentIndex].x
- mLastRawPointerData.pointers[lastIndex].x)
* mPointerXMovementScale;
float deltaY = (mCurrentRawPointerData.pointers[currentIndex].y
- mLastRawPointerData.pointers[lastIndex].y)
* mPointerYMovementScale;
rotateDelta(mSurfaceOrientation, &deltaX, &deltaY);
mPointerVelocityControl.move(when, &deltaX, &deltaY);
mPointerController->move(deltaX, deltaY);
} else {
mPointerVelocityControl.reset();
}
down = isPointerDown(mCurrentButtonState);
hovering = !down;
float x, y;
mPointerController->getPosition(&x, &y);
mPointerSimple.currentCoords.copyFrom(
mCurrentCookedPointerData.pointerCoords[currentIndex]);
mPointerSimple.currentCoords.setAxisValue(AMOTION_EVENT_AXIS_X, x);
mPointerSimple.currentCoords.setAxisValue(AMOTION_EVENT_AXIS_Y, y);
mPointerSimple.currentCoords.setAxisValue(AMOTION_EVENT_AXIS_PRESSURE,
hovering ? 0.0f : 1.0f);
mPointerSimple.currentProperties.id = 0;
mPointerSimple.currentProperties.toolType =
mCurrentCookedPointerData.pointerProperties[currentIndex].toolType;
} else {
mPointerVelocityControl.reset();
down = false;
hovering = false;
}
dispatchPointerSimple(when, policyFlags, down, hovering);
}
void TouchInputMapper::abortPointerMouse(nsecs_t when, uint32_t policyFlags) {
abortPointerSimple(when, policyFlags);
mPointerVelocityControl.reset();
}
void TouchInputMapper::dispatchPointerSimple(nsecs_t when, uint32_t policyFlags,
bool down, bool hovering) {
int32_t metaState = getContext()->getGlobalMetaState();
if (mPointerController != NULL) {
if (down || hovering) {
mPointerController->setPresentation(PointerControllerInterface::PRESENTATION_POINTER);
mPointerController->clearSpots();
mPointerController->setButtonState(mCurrentButtonState);
mPointerController->unfade(PointerControllerInterface::TRANSITION_IMMEDIATE);
} else if (!down && !hovering && (mPointerSimple.down || mPointerSimple.hovering)) {
mPointerController->fade(PointerControllerInterface::TRANSITION_GRADUAL);
}
}
if (mPointerSimple.down && !down) {
mPointerSimple.down = false;
// Send up.
NotifyMotionArgs args(when, getDeviceId(), mSource, policyFlags,
AMOTION_EVENT_ACTION_UP, 0, metaState, mLastButtonState, 0,
1, &mPointerSimple.lastProperties, &mPointerSimple.lastCoords,
mOrientedXPrecision, mOrientedYPrecision,
mPointerSimple.downTime);
getListener()->notifyMotion(&args);
}
if (mPointerSimple.hovering && !hovering) {
mPointerSimple.hovering = false;
// Send hover exit.
NotifyMotionArgs args(when, getDeviceId(), mSource, policyFlags,
AMOTION_EVENT_ACTION_HOVER_EXIT, 0, metaState, mLastButtonState, 0,
1, &mPointerSimple.lastProperties, &mPointerSimple.lastCoords,
mOrientedXPrecision, mOrientedYPrecision,
mPointerSimple.downTime);
getListener()->notifyMotion(&args);
}
if (down) {
if (!mPointerSimple.down) {
mPointerSimple.down = true;
mPointerSimple.downTime = when;
// Send down.
NotifyMotionArgs args(when, getDeviceId(), mSource, policyFlags,
AMOTION_EVENT_ACTION_DOWN, 0, metaState, mCurrentButtonState, 0,
1, &mPointerSimple.currentProperties, &mPointerSimple.currentCoords,
mOrientedXPrecision, mOrientedYPrecision,
mPointerSimple.downTime);
getListener()->notifyMotion(&args);
}
// Send move.
NotifyMotionArgs args(when, getDeviceId(), mSource, policyFlags,
AMOTION_EVENT_ACTION_MOVE, 0, metaState, mCurrentButtonState, 0,
1, &mPointerSimple.currentProperties, &mPointerSimple.currentCoords,
mOrientedXPrecision, mOrientedYPrecision,
mPointerSimple.downTime);
getListener()->notifyMotion(&args);
}
if (hovering) {
if (!mPointerSimple.hovering) {
mPointerSimple.hovering = true;
// Send hover enter.
NotifyMotionArgs args(when, getDeviceId(), mSource, policyFlags,
AMOTION_EVENT_ACTION_HOVER_ENTER, 0, metaState, mCurrentButtonState, 0,
1, &mPointerSimple.currentProperties, &mPointerSimple.currentCoords,
mOrientedXPrecision, mOrientedYPrecision,
mPointerSimple.downTime);
getListener()->notifyMotion(&args);
}
// Send hover move.
NotifyMotionArgs args(when, getDeviceId(), mSource, policyFlags,
AMOTION_EVENT_ACTION_HOVER_MOVE, 0, metaState, mCurrentButtonState, 0,
1, &mPointerSimple.currentProperties, &mPointerSimple.currentCoords,
mOrientedXPrecision, mOrientedYPrecision,
mPointerSimple.downTime);
getListener()->notifyMotion(&args);
}
if (mCurrentRawVScroll || mCurrentRawHScroll) {
float vscroll = mCurrentRawVScroll;
float hscroll = mCurrentRawHScroll;
mWheelYVelocityControl.move(when, NULL, &vscroll);
mWheelXVelocityControl.move(when, &hscroll, NULL);
// Send scroll.
PointerCoords pointerCoords;
pointerCoords.copyFrom(mPointerSimple.currentCoords);
pointerCoords.setAxisValue(AMOTION_EVENT_AXIS_VSCROLL, vscroll);
pointerCoords.setAxisValue(AMOTION_EVENT_AXIS_HSCROLL, hscroll);
NotifyMotionArgs args(when, getDeviceId(), mSource, policyFlags,
AMOTION_EVENT_ACTION_SCROLL, 0, metaState, mCurrentButtonState, 0,
1, &mPointerSimple.currentProperties, &pointerCoords,
mOrientedXPrecision, mOrientedYPrecision,
mPointerSimple.downTime);
getListener()->notifyMotion(&args);
}
// Save state.
if (down || hovering) {
mPointerSimple.lastCoords.copyFrom(mPointerSimple.currentCoords);
mPointerSimple.lastProperties.copyFrom(mPointerSimple.currentProperties);
} else {
mPointerSimple.reset();
}
}
void TouchInputMapper::abortPointerSimple(nsecs_t when, uint32_t policyFlags) {
mPointerSimple.currentCoords.clear();
mPointerSimple.currentProperties.clear();
dispatchPointerSimple(when, policyFlags, false, false);
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
void TouchInputMapper::dispatchMotion(nsecs_t when, uint32_t policyFlags, uint32_t source,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
int32_t action, int32_t flags, int32_t metaState, int32_t buttonState, int32_t edgeFlags,
const PointerProperties* properties, const PointerCoords* coords,
const uint32_t* idToIndex, BitSet32 idBits,
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
int32_t changedId, float xPrecision, float yPrecision, nsecs_t downTime) {
PointerCoords pointerCoords[MAX_POINTERS];
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
PointerProperties pointerProperties[MAX_POINTERS];
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
uint32_t pointerCount = 0;
while (!idBits.isEmpty()) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = idBits.clearFirstMarkedBit();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
uint32_t index = idToIndex[id];
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
pointerProperties[pointerCount].copyFrom(properties[index]);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
pointerCoords[pointerCount].copyFrom(coords[index]);
if (changedId >= 0 && id == uint32_t(changedId)) {
action |= pointerCount << AMOTION_EVENT_ACTION_POINTER_INDEX_SHIFT;
}
pointerCount += 1;
}
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(pointerCount != 0);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
if (changedId >= 0 && pointerCount == 1) {
// Replace initial down and final up action.
// We can compare the action without masking off the changed pointer index
// because we know the index is 0.
if (action == AMOTION_EVENT_ACTION_POINTER_DOWN) {
action = AMOTION_EVENT_ACTION_DOWN;
} else if (action == AMOTION_EVENT_ACTION_POINTER_UP) {
action = AMOTION_EVENT_ACTION_UP;
} else {
// Can't happen.
Rename LOG_ASSERT to ALOG_ASSERT DO NOT MERGE See https://android-git.corp.google.com/g/157519 Bug: 5449033 Change-Id: I8ceb2dba1b031a0fd68d15d146960d9ced62bbf3
2012-01-09 18:35:44 +00:00
ALOG_ASSERT(false);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
NotifyMotionArgs args(when, getDeviceId(), source, policyFlags,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
action, flags, metaState, buttonState, edgeFlags,
pointerCount, pointerProperties, pointerCoords, xPrecision, yPrecision, downTime);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
getListener()->notifyMotion(&args);
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
}
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
bool TouchInputMapper::updateMovedPointers(const PointerProperties* inProperties,
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
const PointerCoords* inCoords, const uint32_t* inIdToIndex,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
PointerProperties* outProperties, PointerCoords* outCoords, const uint32_t* outIdToIndex,
BitSet32 idBits) const {
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
bool changed = false;
while (!idBits.isEmpty()) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = idBits.clearFirstMarkedBit();
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
uint32_t inIndex = inIdToIndex[id];
uint32_t outIndex = outIdToIndex[id];
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
const PointerProperties& curInProperties = inProperties[inIndex];
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
const PointerCoords& curInCoords = inCoords[inIndex];
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
PointerProperties& curOutProperties = outProperties[outIndex];
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
PointerCoords& curOutCoords = outCoords[outIndex];
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
if (curInProperties != curOutProperties) {
curOutProperties.copyFrom(curInProperties);
changed = true;
}
Use touch pad gestures to manipulate the pointer. 1. Single finger tap performs a click. 2. Single finger movement moves the pointer (hovers). 3. Button press plus movement performs click or drag. While dragging, the pointer follows the finger that is moving fastest. This is important if there are additional fingers down on the touch pad for the purpose of applying force to an integrated button underneath. 4. Two fingers near each other moving in the same direction are coalesced as a swipe gesture under the pointer. 5. Two or more fingers moving in arbitrary directions are transformed into touches in the vicinity of the pointer. This makes scale/zoom and rotate gestures possible. Added a native VelocityTracker implementation to enable intelligent switching of the active pointer during drags. Change-Id: I5ada57e7f2bdb9b0a791843eb354a8c706b365dc
2011-03-09 17:39:48 -08:00
if (curInCoords != curOutCoords) {
curOutCoords.copyFrom(curInCoords);
changed = true;
}
}
return changed;
}
void TouchInputMapper::fadePointer() {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (mPointerController != NULL) {
mPointerController->fade(PointerControllerInterface::TRANSITION_GRADUAL);
}
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
bool TouchInputMapper::isPointInsideSurface(int32_t x, int32_t y) {
return x >= mRawPointerAxes.x.minValue && x <= mRawPointerAxes.x.maxValue
&& y >= mRawPointerAxes.y.minValue && y <= mRawPointerAxes.y.maxValue;
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
const TouchInputMapper::VirtualKey* TouchInputMapper::findVirtualKeyHit(
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
int32_t x, int32_t y) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
size_t numVirtualKeys = mVirtualKeys.size();
Improve thread safety of input mappers. Also fixed bug where old touch screen size could be reported by getMotionRange if an orientation change occurred but the user has not yet touched the screen. Bug: 2877345 Change-Id: I7878f47458f310ed6ebe6a5d1b2c9bec2c598ab9
2010-07-29 18:18:33 -07:00
for (size_t i = 0; i < numVirtualKeys; i++) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
const VirtualKey& virtualKey = mVirtualKeys[i];
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#if DEBUG_VIRTUAL_KEYS
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("VirtualKeys: Hit test (%d, %d): keyCode=%d, scanCode=%d, "
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
"left=%d, top=%d, right=%d, bottom=%d",
x, y,
virtualKey.keyCode, virtualKey.scanCode,
virtualKey.hitLeft, virtualKey.hitTop,
virtualKey.hitRight, virtualKey.hitBottom);
#endif
if (virtualKey.isHit(x, y)) {
return & virtualKey;
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
return NULL;
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void TouchInputMapper::assignPointerIds() {
uint32_t currentPointerCount = mCurrentRawPointerData.pointerCount;
uint32_t lastPointerCount = mLastRawPointerData.pointerCount;
mCurrentRawPointerData.clearIdBits();
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
if (currentPointerCount == 0) {
// No pointers to assign.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
return;
}
if (lastPointerCount == 0) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// All pointers are new.
for (uint32_t i = 0; i < currentPointerCount; i++) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = i;
mCurrentRawPointerData.pointers[i].id = id;
mCurrentRawPointerData.idToIndex[id] = i;
mCurrentRawPointerData.markIdBit(id, mCurrentRawPointerData.isHovering(i));
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
return;
}
if (currentPointerCount == 1 && lastPointerCount == 1
&& mCurrentRawPointerData.pointers[0].toolType
== mLastRawPointerData.pointers[0].toolType) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// Only one pointer and no change in count so it must have the same id as before.
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t id = mLastRawPointerData.pointers[0].id;
mCurrentRawPointerData.pointers[0].id = id;
mCurrentRawPointerData.idToIndex[id] = 0;
mCurrentRawPointerData.markIdBit(id, mCurrentRawPointerData.isHovering(0));
return;
}
// General case.
// We build a heap of squared euclidean distances between current and last pointers
// associated with the current and last pointer indices. Then, we find the best
// match (by distance) for each current pointer.
// The pointers must have the same tool type but it is possible for them to
// transition from hovering to touching or vice-versa while retaining the same id.
PointerDistanceHeapElement heap[MAX_POINTERS * MAX_POINTERS];
uint32_t heapSize = 0;
for (uint32_t currentPointerIndex = 0; currentPointerIndex < currentPointerCount;
currentPointerIndex++) {
for (uint32_t lastPointerIndex = 0; lastPointerIndex < lastPointerCount;
lastPointerIndex++) {
const RawPointerData::Pointer& currentPointer =
mCurrentRawPointerData.pointers[currentPointerIndex];
const RawPointerData::Pointer& lastPointer =
mLastRawPointerData.pointers[lastPointerIndex];
if (currentPointer.toolType == lastPointer.toolType) {
int64_t deltaX = currentPointer.x - lastPointer.x;
int64_t deltaY = currentPointer.y - lastPointer.y;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
uint64_t distance = uint64_t(deltaX * deltaX + deltaY * deltaY);
// Insert new element into the heap (sift up).
heap[heapSize].currentPointerIndex = currentPointerIndex;
heap[heapSize].lastPointerIndex = lastPointerIndex;
heap[heapSize].distance = distance;
heapSize += 1;
}
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// Heapify
for (uint32_t startIndex = heapSize / 2; startIndex != 0; ) {
startIndex -= 1;
for (uint32_t parentIndex = startIndex; ;) {
uint32_t childIndex = parentIndex * 2 + 1;
if (childIndex >= heapSize) {
break;
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (childIndex + 1 < heapSize
&& heap[childIndex + 1].distance < heap[childIndex].distance) {
childIndex += 1;
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (heap[parentIndex].distance <= heap[childIndex].distance) {
break;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
swap(heap[parentIndex], heap[childIndex]);
parentIndex = childIndex;
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#if DEBUG_POINTER_ASSIGNMENT
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("assignPointerIds - initial distance min-heap: size=%d", heapSize);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
for (size_t i = 0; i < heapSize; i++) {
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD(" heap[%d]: cur=%d, last=%d, distance=%lld",
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
i, heap[i].currentPointerIndex, heap[i].lastPointerIndex,
heap[i].distance);
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#endif
More work in progress on native events. Refactored the code to eliminate potential deadlocks due to re-entrant calls from the policy into the dispatcher. Also added some plumbing that will be used to notify the framework about ANRs. Change-Id: Iba7a10de0cb3c56cd7520d6ce716db52fdcc94ff
2010-06-15 01:31:58 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// Pull matches out by increasing order of distance.
// To avoid reassigning pointers that have already been matched, the loop keeps track
// of which last and current pointers have been matched using the matchedXXXBits variables.
// It also tracks the used pointer id bits.
BitSet32 matchedLastBits(0);
BitSet32 matchedCurrentBits(0);
BitSet32 usedIdBits(0);
bool first = true;
for (uint32_t i = min(currentPointerCount, lastPointerCount); heapSize > 0 && i > 0; i--) {
while (heapSize > 0) {
if (first) {
// The first time through the loop, we just consume the root element of
// the heap (the one with smallest distance).
first = false;
} else {
// Previous iterations consumed the root element of the heap.
// Pop root element off of the heap (sift down).
heap[0] = heap[heapSize];
for (uint32_t parentIndex = 0; ;) {
uint32_t childIndex = parentIndex * 2 + 1;
if (childIndex >= heapSize) {
break;
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (childIndex + 1 < heapSize
&& heap[childIndex + 1].distance < heap[childIndex].distance) {
childIndex += 1;
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (heap[parentIndex].distance <= heap[childIndex].distance) {
break;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
swap(heap[parentIndex], heap[childIndex]);
parentIndex = childIndex;
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#if DEBUG_POINTER_ASSIGNMENT
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("assignPointerIds - reduced distance min-heap: size=%d", heapSize);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
for (size_t i = 0; i < heapSize; i++) {
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD(" heap[%d]: cur=%d, last=%d, distance=%lld",
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
i, heap[i].currentPointerIndex, heap[i].lastPointerIndex,
heap[i].distance);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
#endif
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
heapSize -= 1;
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t currentPointerIndex = heap[0].currentPointerIndex;
if (matchedCurrentBits.hasBit(currentPointerIndex)) continue; // already matched
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t lastPointerIndex = heap[0].lastPointerIndex;
if (matchedLastBits.hasBit(lastPointerIndex)) continue; // already matched
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
matchedCurrentBits.markBit(currentPointerIndex);
matchedLastBits.markBit(lastPointerIndex);
uint32_t id = mLastRawPointerData.pointers[lastPointerIndex].id;
mCurrentRawPointerData.pointers[currentPointerIndex].id = id;
mCurrentRawPointerData.idToIndex[id] = currentPointerIndex;
mCurrentRawPointerData.markIdBit(id,
mCurrentRawPointerData.isHovering(currentPointerIndex));
usedIdBits.markBit(id);
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#if DEBUG_POINTER_ASSIGNMENT
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("assignPointerIds - matched: cur=%d, last=%d, id=%d, distance=%lld",
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
lastPointerIndex, currentPointerIndex, id, heap[0].distance);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#endif
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
break;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
// Assign fresh ids to pointers that were not matched in the process.
for (uint32_t i = currentPointerCount - matchedCurrentBits.count(); i != 0; i--) {
uint32_t currentPointerIndex = matchedCurrentBits.markFirstUnmarkedBit();
uint32_t id = usedIdBits.markFirstUnmarkedBit();
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentRawPointerData.pointers[currentPointerIndex].id = id;
mCurrentRawPointerData.idToIndex[id] = currentPointerIndex;
mCurrentRawPointerData.markIdBit(id,
mCurrentRawPointerData.isHovering(currentPointerIndex));
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#if DEBUG_POINTER_ASSIGNMENT
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("assignPointerIds - assigned: cur=%d, id=%d",
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
currentPointerIndex, id);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
#endif
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
int32_t TouchInputMapper::getKeyCodeState(uint32_t sourceMask, int32_t keyCode) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (mCurrentVirtualKey.down && mCurrentVirtualKey.keyCode == keyCode) {
return AKEY_STATE_VIRTUAL;
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
size_t numVirtualKeys = mVirtualKeys.size();
for (size_t i = 0; i < numVirtualKeys; i++) {
const VirtualKey& virtualKey = mVirtualKeys[i];
if (virtualKey.keyCode == keyCode) {
return AKEY_STATE_UP;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
return AKEY_STATE_UNKNOWN;
}
int32_t TouchInputMapper::getScanCodeState(uint32_t sourceMask, int32_t scanCode) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
if (mCurrentVirtualKey.down && mCurrentVirtualKey.scanCode == scanCode) {
return AKEY_STATE_VIRTUAL;
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
size_t numVirtualKeys = mVirtualKeys.size();
for (size_t i = 0; i < numVirtualKeys; i++) {
const VirtualKey& virtualKey = mVirtualKeys[i];
if (virtualKey.scanCode == scanCode) {
return AKEY_STATE_UP;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
return AKEY_STATE_UNKNOWN;
}
bool TouchInputMapper::markSupportedKeyCodes(uint32_t sourceMask, size_t numCodes,
const int32_t* keyCodes, uint8_t* outFlags) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
size_t numVirtualKeys = mVirtualKeys.size();
for (size_t i = 0; i < numVirtualKeys; i++) {
const VirtualKey& virtualKey = mVirtualKeys[i];
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
for (size_t i = 0; i < numCodes; i++) {
if (virtualKey.keyCode == keyCodes[i]) {
outFlags[i] = 1;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
return true;
}
// --- SingleTouchInputMapper ---
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
SingleTouchInputMapper::SingleTouchInputMapper(InputDevice* device) :
TouchInputMapper(device) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
SingleTouchInputMapper::~SingleTouchInputMapper() {
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void SingleTouchInputMapper::reset(nsecs_t when) {
mSingleTouchMotionAccumulator.reset(getDevice());
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
TouchInputMapper::reset(when);
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
void SingleTouchInputMapper::process(const RawEvent* rawEvent) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
TouchInputMapper::process(rawEvent);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mSingleTouchMotionAccumulator.process(rawEvent);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void SingleTouchInputMapper::syncTouch(nsecs_t when, bool* outHavePointerIds) {
Use BTN_TOUCH or BTN_TOOL_* to determine if touch active. Bug: 5064702 Should not include stylus buttons in the condition. Change-Id: If4d78a875b77da8bd59672d6fdbf5353004d0023
2011-08-10 14:55:59 -07:00
if (mTouchButtonAccumulator.isToolActive()) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentRawPointerData.pointerCount = 1;
mCurrentRawPointerData.idToIndex[0] = 0;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
bool isHovering = mTouchButtonAccumulator.getToolType() != AMOTION_EVENT_TOOL_TYPE_MOUSE
&& (mTouchButtonAccumulator.isHovering()
|| (mRawPointerAxes.pressure.valid
&& mSingleTouchMotionAccumulator.getAbsolutePressure() <= 0));
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentRawPointerData.markIdBit(0, isHovering);
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
RawPointerData::Pointer& outPointer = mCurrentRawPointerData.pointers[0];
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
outPointer.id = 0;
outPointer.x = mSingleTouchMotionAccumulator.getAbsoluteX();
outPointer.y = mSingleTouchMotionAccumulator.getAbsoluteY();
outPointer.pressure = mSingleTouchMotionAccumulator.getAbsolutePressure();
outPointer.touchMajor = 0;
outPointer.touchMinor = 0;
outPointer.toolMajor = mSingleTouchMotionAccumulator.getAbsoluteToolWidth();
outPointer.toolMinor = mSingleTouchMotionAccumulator.getAbsoluteToolWidth();
outPointer.orientation = 0;
outPointer.distance = mSingleTouchMotionAccumulator.getAbsoluteDistance();
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
outPointer.tiltX = mSingleTouchMotionAccumulator.getAbsoluteTiltX();
outPointer.tiltY = mSingleTouchMotionAccumulator.getAbsoluteTiltY();
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
outPointer.toolType = mTouchButtonAccumulator.getToolType();
if (outPointer.toolType == AMOTION_EVENT_TOOL_TYPE_UNKNOWN) {
outPointer.toolType = AMOTION_EVENT_TOOL_TYPE_FINGER;
}
outPointer.isHovering = isHovering;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void SingleTouchInputMapper::configureRawPointerAxes() {
TouchInputMapper::configureRawPointerAxes();
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
getAbsoluteAxisInfo(ABS_X, &mRawPointerAxes.x);
getAbsoluteAxisInfo(ABS_Y, &mRawPointerAxes.y);
getAbsoluteAxisInfo(ABS_PRESSURE, &mRawPointerAxes.pressure);
getAbsoluteAxisInfo(ABS_TOOL_WIDTH, &mRawPointerAxes.toolMajor);
getAbsoluteAxisInfo(ABS_DISTANCE, &mRawPointerAxes.distance);
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
getAbsoluteAxisInfo(ABS_TILT_X, &mRawPointerAxes.tiltX);
getAbsoluteAxisInfo(ABS_TILT_Y, &mRawPointerAxes.tiltY);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
bool SingleTouchInputMapper::hasStylus() const {
return mTouchButtonAccumulator.hasStylus();
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
// --- MultiTouchInputMapper ---
Support non-orientation aware keyboards and other devices. Fixed a bug with dpad keys on external keyboards being rotated according to the display orientation by adding a new input device configuration property called "keyboard.orientationAware". Added a mechanism for overriding the key layout and key character map in the input device configuration file using the new "keyboard.layout" and "keyboard.characterMap" properties. Also added "trackball.orientationAware", "touch.orientationAware" and "touch.deviceType" configuration properties. Rewrote the configuration property reading code in native code so that it can be used by EventHub and other components. Added basic support for installable idc, kl, and kcm files in /data/system/devices. However, there is no provision for copying files there yet. Disabled long-press character pickers on full keyboards so that key repeating works as expected. Change-Id: I1bd9f0c3d344421db444e7d271eb09bc8bab4791
2010-11-29 17:37:49 -08:00
MultiTouchInputMapper::MultiTouchInputMapper(InputDevice* device) :
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
TouchInputMapper(device) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
MultiTouchInputMapper::~MultiTouchInputMapper() {
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void MultiTouchInputMapper::reset(nsecs_t when) {
mMultiTouchMotionAccumulator.reset(getDevice());
Query input device for initial slot index. This fixes a problem where touches can get stuck because the driver and the framework have different ideas of what the initial slot index is. The framework assumed it was slot 0 but it could in principle be any slot, such as slot 1. When that happened, the framework would start tracking the first touch as slot 0, but it might never receive an "up" for that slot. Change-Id: Idaffc4534b275d66b9d4360987b28dc2d0f63218
2011-06-30 23:53:07 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mPointerIdBits.clear();
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
TouchInputMapper::reset(when);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
void MultiTouchInputMapper::process(const RawEvent* rawEvent) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
TouchInputMapper::process(rawEvent);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mMultiTouchMotionAccumulator.process(rawEvent);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void MultiTouchInputMapper::syncTouch(nsecs_t when, bool* outHavePointerIds) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
size_t inCount = mMultiTouchMotionAccumulator.getSlotCount();
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
size_t outCount = 0;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
BitSet32 newPointerIdBits;
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
for (size_t inIndex = 0; inIndex < inCount; inIndex++) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
const MultiTouchMotionAccumulator::Slot* inSlot =
mMultiTouchMotionAccumulator.getSlot(inIndex);
if (!inSlot->isInUse()) {
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
continue;
}
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
if (outCount >= MAX_POINTERS) {
#if DEBUG_POINTERS
Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156016 Bug: 5449033 Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2011-12-20 16:23:08 +00:00
ALOGD("MultiTouch device %s emitted more than maximum of %d pointers; "
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
"ignoring the rest.",
getDeviceName().string(), MAX_POINTERS);
#endif
break; // too many fingers!
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
RawPointerData::Pointer& outPointer = mCurrentRawPointerData.pointers[outCount];
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
outPointer.x = inSlot->getX();
outPointer.y = inSlot->getY();
outPointer.pressure = inSlot->getPressure();
outPointer.touchMajor = inSlot->getTouchMajor();
outPointer.touchMinor = inSlot->getTouchMinor();
outPointer.toolMajor = inSlot->getToolMajor();
outPointer.toolMinor = inSlot->getToolMinor();
outPointer.orientation = inSlot->getOrientation();
outPointer.distance = inSlot->getDistance();
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
outPointer.tiltX = 0;
outPointer.tiltY = 0;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
outPointer.toolType = inSlot->getToolType();
if (outPointer.toolType == AMOTION_EVENT_TOOL_TYPE_UNKNOWN) {
outPointer.toolType = mTouchButtonAccumulator.getToolType();
if (outPointer.toolType == AMOTION_EVENT_TOOL_TYPE_UNKNOWN) {
outPointer.toolType = AMOTION_EVENT_TOOL_TYPE_FINGER;
}
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
bool isHovering = mTouchButtonAccumulator.getToolType() != AMOTION_EVENT_TOOL_TYPE_MOUSE
&& (mTouchButtonAccumulator.isHovering()
|| (mRawPointerAxes.pressure.valid && inSlot->getPressure() <= 0));
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
outPointer.isHovering = isHovering;
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
Input device calibration and capabilities. Finished the input device capability API. Added a mechanism for calibrating touch devices to obtain more accurate information about the touch contact area. Improved pointer location to show new coordinates and capabilities. Optimized pointer location display and formatting to avoid allocating large numbers of temporary objects. The GC churn was causing the application to stutter very badly when more than a couple of fingers were down). Added more diagnostics. Change-Id: Ie25380278ed6f16c5b04cd9df848015850383498
2010-08-30 03:02:23 -07:00
// Assign pointer id using tracking id if available.
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
if (*outHavePointerIds) {
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
int32_t trackingId = inSlot->getTrackingId();
Workaround apps that make assumptions about pointer ids. Modified the touch input mapper to assign pointer ids sequentially starting from 0 instead of using the tracking id or slot index supplied by the driver. Applications should not depend on this ordering but some do. (sigh) Bug: 4980884 Change-Id: I0dfeb3ac27c57a7102a13c960c760e2a02eb7669
2011-07-01 17:59:27 -07:00
int32_t id = -1;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
if (trackingId >= 0) {
Workaround apps that make assumptions about pointer ids. Modified the touch input mapper to assign pointer ids sequentially starting from 0 instead of using the tracking id or slot index supplied by the driver. Applications should not depend on this ordering but some do. (sigh) Bug: 4980884 Change-Id: I0dfeb3ac27c57a7102a13c960c760e2a02eb7669
2011-07-01 17:59:27 -07:00
for (BitSet32 idBits(mPointerIdBits); !idBits.isEmpty(); ) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
uint32_t n = idBits.clearFirstMarkedBit();
Workaround apps that make assumptions about pointer ids. Modified the touch input mapper to assign pointer ids sequentially starting from 0 instead of using the tracking id or slot index supplied by the driver. Applications should not depend on this ordering but some do. (sigh) Bug: 4980884 Change-Id: I0dfeb3ac27c57a7102a13c960c760e2a02eb7669
2011-07-01 17:59:27 -07:00
if (mPointerTrackingIdMap[n] == trackingId) {
id = n;
}
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Workaround apps that make assumptions about pointer ids. Modified the touch input mapper to assign pointer ids sequentially starting from 0 instead of using the tracking id or slot index supplied by the driver. Applications should not depend on this ordering but some do. (sigh) Bug: 4980884 Change-Id: I0dfeb3ac27c57a7102a13c960c760e2a02eb7669
2011-07-01 17:59:27 -07:00
if (id < 0 && !mPointerIdBits.isFull()) {
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
id = mPointerIdBits.markFirstUnmarkedBit();
Workaround apps that make assumptions about pointer ids. Modified the touch input mapper to assign pointer ids sequentially starting from 0 instead of using the tracking id or slot index supplied by the driver. Applications should not depend on this ordering but some do. (sigh) Bug: 4980884 Change-Id: I0dfeb3ac27c57a7102a13c960c760e2a02eb7669
2011-07-01 17:59:27 -07:00
mPointerTrackingIdMap[id] = trackingId;
}
}
if (id < 0) {
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
*outHavePointerIds = false;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentRawPointerData.clearIdBits();
newPointerIdBits.clear();
Workaround apps that make assumptions about pointer ids. Modified the touch input mapper to assign pointer ids sequentially starting from 0 instead of using the tracking id or slot index supplied by the driver. Applications should not depend on this ordering but some do. (sigh) Bug: 4980884 Change-Id: I0dfeb3ac27c57a7102a13c960c760e2a02eb7669
2011-07-01 17:59:27 -07:00
} else {
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
outPointer.id = id;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentRawPointerData.idToIndex[id] = outCount;
mCurrentRawPointerData.markIdBit(id, isHovering);
newPointerIdBits.markBit(id);
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
}
outCount += 1;
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
mCurrentRawPointerData.pointerCount = outCount;
mPointerIdBits = newPointerIdBits;
Workaround apps that make assumptions about pointer ids. Modified the touch input mapper to assign pointer ids sequentially starting from 0 instead of using the tracking id or slot index supplied by the driver. Applications should not depend on this ordering but some do. (sigh) Bug: 4980884 Change-Id: I0dfeb3ac27c57a7102a13c960c760e2a02eb7669
2011-07-01 17:59:27 -07:00
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
mMultiTouchMotionAccumulator.finishSync();
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
void MultiTouchInputMapper::configureRawPointerAxes() {
TouchInputMapper::configureRawPointerAxes();
Fix some input device mapping bugs with certain drivers. On single-touch devices, pointer up/down is signalled by a BTN_TOUCH key event. Previously we handled BTN_TOUCH immediately but some drivers may produce the sequence BTN_TOUCH, ABS_X, ABS_Y, SYN_REPORT on pointer down which caused us to emit a bad initial pointer down location. Now we wait for SYN_REPORT before reporting the up or down. On multi-touch devices, pointer up can be signalled by as little as the sequence SYN_MT_REPORT, SYN_REPORT. This change ensures that we handle this case. Added support for reading ABS_MT_PRESSURE when available. Corrected mapping of touchMajor/touchMinor on single touch devices. Minor code cleanup. Change-Id: Ic7ec4811241ed85a06e59b8a839ca05180d491d4
2010-08-17 20:38:35 -07:00
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
getAbsoluteAxisInfo(ABS_MT_POSITION_X, &mRawPointerAxes.x);
getAbsoluteAxisInfo(ABS_MT_POSITION_Y, &mRawPointerAxes.y);
getAbsoluteAxisInfo(ABS_MT_TOUCH_MAJOR, &mRawPointerAxes.touchMajor);
getAbsoluteAxisInfo(ABS_MT_TOUCH_MINOR, &mRawPointerAxes.touchMinor);
getAbsoluteAxisInfo(ABS_MT_WIDTH_MAJOR, &mRawPointerAxes.toolMajor);
getAbsoluteAxisInfo(ABS_MT_WIDTH_MINOR, &mRawPointerAxes.toolMinor);
getAbsoluteAxisInfo(ABS_MT_ORIENTATION, &mRawPointerAxes.orientation);
getAbsoluteAxisInfo(ABS_MT_PRESSURE, &mRawPointerAxes.pressure);
getAbsoluteAxisInfo(ABS_MT_DISTANCE, &mRawPointerAxes.distance);
getAbsoluteAxisInfo(ABS_MT_TRACKING_ID, &mRawPointerAxes.trackingId);
getAbsoluteAxisInfo(ABS_MT_SLOT, &mRawPointerAxes.slot);
if (mRawPointerAxes.trackingId.valid
&& mRawPointerAxes.slot.valid
&& mRawPointerAxes.slot.minValue == 0 && mRawPointerAxes.slot.maxValue > 0) {
size_t slotCount = mRawPointerAxes.slot.maxValue + 1;
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
if (slotCount > MAX_SLOTS) {
Rename (IF_)LOGW(_IF) to (IF_)ALOGW(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/157065 Bug: 5449033 Change-Id: I00a4b904f9449e6f93b7fd35eac28640d7929e69
2012-01-05 23:22:43 +00:00
ALOGW("MultiTouch Device %s reported %d slots but the framework "
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
"only supports a maximum of %d slots at this time.",
Handle stylus buttons and tool types. Added TOOL_TYPE_ERASER. Refactored the InputReader to share more code between the various input mappers that handle button states and to simplify the accumulator implementations by having each one only handle a single type of input. Removed the concept of direct/indirect tool types from the API. If we add it back, it should be done in a manner that is orthogonal to the tool type itself, perhaps as a flags field on the pointer. The device source may well provide sufficient information anyhow. Change-Id: I811c22d95e8304269b6ee4f6d11a6b04f3cfc1b2
2011-07-01 17:37:58 -07:00
getDeviceName().string(), slotCount, MAX_SLOTS);
slotCount = MAX_SLOTS;
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
}
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
mMultiTouchMotionAccumulator.configure(getDevice(),
slotCount, true /*usingSlotsProtocol*/);
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
} else {
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
mMultiTouchMotionAccumulator.configure(getDevice(),
MAX_POINTERS, false /*usingSlotsProtocol*/);
Input device protocol enhancements. Added support for Linux multitouch protocol B (slots). Added support for using the device's input properties as a hint to determine the intended usage of a touch device. Added support for the ABS_MT_DISTANCE axis. Fixed a bug reporting the presence of the orientation axis. Change-Id: Icf7b5a5a0f1a9cdf6ad2b35be8ea0c1a35815d48
2011-05-24 01:07:44 -07:00
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
}
Make InputDevice.SOURCE_STYLUS meaningful. Bug: 5424551 Change-Id: I415cb1842422e050cc41b17f5a1f13b4fab17a44
2012-04-19 15:18:26 -07:00
bool MultiTouchInputMapper::hasStylus() const {
return mMultiTouchMotionAccumulator.hasStylus()
|| mTouchButtonAccumulator.hasStylus();
}
Refactor input reader to support new device types more easily. Refactored the input reader so that each raw input protocol is handled by a separate subclass of the new InputMapper type. This way, behaviors pertaining to keyboard, trackballs, touchscreens, switches and other devices are clearly distinguished for improved maintainability. Added partial support for describing capabilities of input devices (incomplete and untested for now, will be fleshed out in later commits). Simplified EventHub interface somewhat since InputReader is taking over more of the work. Cleaned up some of the interactions between InputManager and WindowManagerService related to reading input state. Fixed swiping finger from screen edge into display area. Added logging of device information to 'dumpsys window'. Change-Id: I17faffc33e3aec3a0f33f0b37e81a70609378612
2010-07-23 21:28:06 -07:00
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
// --- JoystickInputMapper ---
JoystickInputMapper::JoystickInputMapper(InputDevice* device) :
InputMapper(device) {
}
JoystickInputMapper::~JoystickInputMapper() {
}
uint32_t JoystickInputMapper::getSources() {
return AINPUT_SOURCE_JOYSTICK;
}
void JoystickInputMapper::populateDeviceInfo(InputDeviceInfo* info) {
InputMapper::populateDeviceInfo(info);
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
for (size_t i = 0; i < mAxes.size(); i++) {
const Axis& axis = mAxes.valueAt(i);
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
info->addMotionRange(axis.axisInfo.axis, AINPUT_SOURCE_JOYSTICK,
axis.min, axis.max, axis.flat, axis.fuzz);
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
if (axis.axisInfo.mode == AxisInfo::MODE_SPLIT) {
Input improvements and bug fixes. Associate each motion axis with the source from which it comes. It is possible for multiple sources of the same device to define the same axis. This fixes new API that was introduced in MR1. (Bug: 4066146) Fixed a bug that might cause a segfault when using a trackball. Only fade out the mouse pointer when touching the touch screen, ignore other touch pads. Changed the plural "sources" to "source" in several places in the InputReader where we intend to refer to a particular source rather than to a combination of sources. Improved the batching code to support batching events from different sources of the same device in parallel. (Bug: 3391564) Change-Id: I0189e18e464338f126f7bf94370b928e1b1695f2
2011-03-08 15:13:06 -08:00
info->addMotionRange(axis.axisInfo.highAxis, AINPUT_SOURCE_JOYSTICK,
axis.min, axis.max, axis.flat, axis.fuzz);
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
}
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
}
}
void JoystickInputMapper::dump(String8& dump) {
dump.append(INDENT2 "Joystick Input Mapper:\n");
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
dump.append(INDENT3 "Axes:\n");
size_t numAxes = mAxes.size();
for (size_t i = 0; i < numAxes; i++) {
const Axis& axis = mAxes.valueAt(i);
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
const char* label = getAxisLabel(axis.axisInfo.axis);
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
if (label) {
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
dump.appendFormat(INDENT4 "%s", label);
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
} else {
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
dump.appendFormat(INDENT4 "%d", axis.axisInfo.axis);
}
if (axis.axisInfo.mode == AxisInfo::MODE_SPLIT) {
label = getAxisLabel(axis.axisInfo.highAxis);
if (label) {
dump.appendFormat(" / %s (split at %d)", label, axis.axisInfo.splitValue);
} else {
dump.appendFormat(" / %d (split at %d)", axis.axisInfo.highAxis,
axis.axisInfo.splitValue);
}
} else if (axis.axisInfo.mode == AxisInfo::MODE_INVERT) {
dump.append(" (invert)");
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
}
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
dump.appendFormat(": min=%0.5f, max=%0.5f, flat=%0.5f, fuzz=%0.5f\n",
axis.min, axis.max, axis.flat, axis.fuzz);
dump.appendFormat(INDENT4 " scale=%0.5f, offset=%0.5f, "
"highScale=%0.5f, highOffset=%0.5f\n",
axis.scale, axis.offset, axis.highScale, axis.highOffset);
Take advantage of updated linux/input.h kernel headers. Change-Id: I72d2ef82de5c504d46b0cdb57aa43bbd0d769174
2011-06-12 18:14:50 -07:00
dump.appendFormat(INDENT4 " rawAxis=%d, rawMin=%d, rawMax=%d, "
"rawFlat=%d, rawFuzz=%d, rawResolution=%d\n",
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
mAxes.keyAt(i), axis.rawAxisInfo.minValue, axis.rawAxisInfo.maxValue,
Take advantage of updated linux/input.h kernel headers. Change-Id: I72d2ef82de5c504d46b0cdb57aa43bbd0d769174
2011-06-12 18:14:50 -07:00
axis.rawAxisInfo.flat, axis.rawAxisInfo.fuzz, axis.rawAxisInfo.resolution);
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
}
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void JoystickInputMapper::configure(nsecs_t when,
const InputReaderConfiguration* config, uint32_t changes) {
InputMapper::configure(when, config, changes);
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (!changes) { // first time only
// Collect all axes.
for (int32_t abs = 0; abs <= ABS_MAX; abs++) {
Support composite touch / joystick devices better. This change enables the joystick input mapper to handle any axes that are not claimed by the touch input mapper, which makes auxiliary controls such as wheels / knobs accessible. Change-Id: I01ee7f342ac91acfcb4ccb6676fd52b3d5bf31a0
2011-08-31 12:56:34 -07:00
if (!(getAbsAxisUsage(abs, getDevice()->getClasses())
& INPUT_DEVICE_CLASS_JOYSTICK)) {
continue; // axis must be claimed by a different device
}
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
RawAbsoluteAxisInfo rawAxisInfo;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
getAbsoluteAxisInfo(abs, &rawAxisInfo);
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (rawAxisInfo.valid) {
// Map axis.
AxisInfo axisInfo;
bool explicitlyMapped = !getEventHub()->mapAxis(getDeviceId(), abs, &axisInfo);
if (!explicitlyMapped) {
// Axis is not explicitly mapped, will choose a generic axis later.
axisInfo.mode = AxisInfo::MODE_NORMAL;
axisInfo.axis = -1;
}
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
// Apply flat override.
int32_t rawFlat = axisInfo.flatOverride < 0
? rawAxisInfo.flat : axisInfo.flatOverride;
// Calculate scaling factors and limits.
Axis axis;
if (axisInfo.mode == AxisInfo::MODE_SPLIT) {
float scale = 1.0f / (axisInfo.splitValue - rawAxisInfo.minValue);
float highScale = 1.0f / (rawAxisInfo.maxValue - axisInfo.splitValue);
axis.initialize(rawAxisInfo, axisInfo, explicitlyMapped,
scale, 0.0f, highScale, 0.0f,
0.0f, 1.0f, rawFlat * scale, rawAxisInfo.fuzz * scale);
} else if (isCenteredAxis(axisInfo.axis)) {
float scale = 2.0f / (rawAxisInfo.maxValue - rawAxisInfo.minValue);
float offset = avg(rawAxisInfo.minValue, rawAxisInfo.maxValue) * -scale;
axis.initialize(rawAxisInfo, axisInfo, explicitlyMapped,
scale, offset, scale, offset,
-1.0f, 1.0f, rawFlat * scale, rawAxisInfo.fuzz * scale);
} else {
float scale = 1.0f / (rawAxisInfo.maxValue - rawAxisInfo.minValue);
axis.initialize(rawAxisInfo, axisInfo, explicitlyMapped,
scale, 0.0f, scale, 0.0f,
0.0f, 1.0f, rawFlat * scale, rawAxisInfo.fuzz * scale);
}
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
// To eliminate noise while the joystick is at rest, filter out small variations
// in axis values up front.
axis.filter = axis.flat * 0.25f;
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
mAxes.add(abs, axis);
}
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
}
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
// If there are too many axes, start dropping them.
// Prefer to keep explicitly mapped axes.
if (mAxes.size() > PointerCoords::MAX_AXES) {
Rename (IF_)LOGI(_IF) to (IF_)ALOGI(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156801 Bug: 5449033 Change-Id: Ib08fe86d23db91ee153e9f91a99a35c42b9208ea
2012-01-04 20:05:49 +00:00
ALOGI("Joystick '%s' has %d axes but the framework only supports a maximum of %d.",
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
getDeviceName().string(), mAxes.size(), PointerCoords::MAX_AXES);
pruneAxes(true);
pruneAxes(false);
}
// Assign generic axis ids to remaining axes.
int32_t nextGenericAxisId = AMOTION_EVENT_AXIS_GENERIC_1;
size_t numAxes = mAxes.size();
for (size_t i = 0; i < numAxes; i++) {
Axis& axis = mAxes.editValueAt(i);
if (axis.axisInfo.axis < 0) {
while (nextGenericAxisId <= AMOTION_EVENT_AXIS_GENERIC_16
&& haveAxis(nextGenericAxisId)) {
nextGenericAxisId += 1;
}
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
if (nextGenericAxisId <= AMOTION_EVENT_AXIS_GENERIC_16) {
axis.axisInfo.axis = nextGenericAxisId;
nextGenericAxisId += 1;
} else {
Rename (IF_)LOGI(_IF) to (IF_)ALOGI(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156801 Bug: 5449033 Change-Id: Ib08fe86d23db91ee153e9f91a99a35c42b9208ea
2012-01-04 20:05:49 +00:00
ALOGI("Ignoring joystick '%s' axis %d because all of the generic axis ids "
Add support for disabling pointer gestures. Made it possible for individual windows to disable pointer gestures while the window has focus using a private API. Cleaned up the InputReader configuration code to enable in-place reconfiguration of input devices without having to reopen them all. This change makes changing the pointer speed somewhat nicer since the pointer doesn't jump back to the origin after each change. Change-Id: I9727419c2f4cb39e16acb4b15fd7fd84526b1239
2011-06-14 20:22:50 -07:00
"have already been assigned to other axes.",
getDeviceName().string(), mAxes.keyAt(i));
mAxes.removeItemsAt(i--);
numAxes -= 1;
}
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
}
}
}
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
}
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
bool JoystickInputMapper::haveAxis(int32_t axisId) {
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
size_t numAxes = mAxes.size();
for (size_t i = 0; i < numAxes; i++) {
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
const Axis& axis = mAxes.valueAt(i);
if (axis.axisInfo.axis == axisId
|| (axis.axisInfo.mode == AxisInfo::MODE_SPLIT
&& axis.axisInfo.highAxis == axisId)) {
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
return true;
}
}
return false;
}
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
void JoystickInputMapper::pruneAxes(bool ignoreExplicitlyMappedAxes) {
size_t i = mAxes.size();
while (mAxes.size() > PointerCoords::MAX_AXES && i-- > 0) {
if (ignoreExplicitlyMappedAxes && mAxes.valueAt(i).explicitlyMapped) {
continue;
}
Rename (IF_)LOGI(_IF) to (IF_)ALOGI(_IF) DO NOT MERGE See https://android-git.corp.google.com/g/156801 Bug: 5449033 Change-Id: Ib08fe86d23db91ee153e9f91a99a35c42b9208ea
2012-01-04 20:05:49 +00:00
ALOGI("Discarding joystick '%s' axis %d because there are too many axes.",
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
getDeviceName().string(), mAxes.keyAt(i));
mAxes.removeItemsAt(i);
}
}
bool JoystickInputMapper::isCenteredAxis(int32_t axis) {
switch (axis) {
case AMOTION_EVENT_AXIS_X:
case AMOTION_EVENT_AXIS_Y:
case AMOTION_EVENT_AXIS_Z:
case AMOTION_EVENT_AXIS_RX:
case AMOTION_EVENT_AXIS_RY:
case AMOTION_EVENT_AXIS_RZ:
case AMOTION_EVENT_AXIS_HAT_X:
case AMOTION_EVENT_AXIS_HAT_Y:
case AMOTION_EVENT_AXIS_ORIENTATION:
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
case AMOTION_EVENT_AXIS_RUDDER:
case AMOTION_EVENT_AXIS_WHEEL:
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
return true;
default:
return false;
}
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
void JoystickInputMapper::reset(nsecs_t when) {
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
// Recenter all axes.
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
size_t numAxes = mAxes.size();
for (size_t i = 0; i < numAxes; i++) {
Axis& axis = mAxes.editValueAt(i);
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
axis.resetValue();
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
}
Input system bug fixes, particularly for stylus. Bug: 5049148 Finished stylus support, including support for indirect stylus and mouse tools. Added TILT axis. When stylus tilt X/Y is available, it is transformed into an orientation and tilt inclination which is a more convenient representation and a simpler extension to the exiting API. Touch devices now only report touch data using a single input source. Previously touch devices in pointer mode would report both absolute touch pad data and cooked pointer gestures. Now we just pick one. The touch device switches modes as needed when the focused application enables/disables pointer gestures. This change greatly simplifies the code and reduces the load on the input dispatcher. Fixed an incorrect assumption that the value of ABS_(MT_)DISTANCE would be zero whenever the stylus was in direct contact. It appears that the correct way to determine whether the stylus is in direct contact (rather than hovering) is by checking for a non-zero reported pressure. Added code to read the initial state of tool buttons and axis values when the input devices are initialized or reset. This fixes problems where the input mapper state might have the wrong initial state. Moved responsibility for cancelling pending inputs (keys down, touches, etc.) to the InputDispatcher by sending it a device reset notification. This frees the InputReader from having to synthesize events during reset, which was cumbersome and somewhat brittle to begin with. Consolidated more of the common accumulator logic from SingleTouchInputMapper and MultiTouchInputMapper into TouchInputMapper. Improved the PointerLocation output. Change-Id: I595d3647f7fd7cb1e3eff8b3c76b85043b5fe2f0
2011-08-18 11:20:58 -07:00
InputMapper::reset(when);
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
}
void JoystickInputMapper::process(const RawEvent* rawEvent) {
switch (rawEvent->type) {
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
case EV_ABS: {
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
ssize_t index = mAxes.indexOfKey(rawEvent->code);
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
if (index >= 0) {
Axis& axis = mAxes.editValueAt(index);
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
float newValue, highNewValue;
switch (axis.axisInfo.mode) {
case AxisInfo::MODE_INVERT:
newValue = (axis.rawAxisInfo.maxValue - rawEvent->value)
* axis.scale + axis.offset;
highNewValue = 0.0f;
break;
case AxisInfo::MODE_SPLIT:
if (rawEvent->value < axis.axisInfo.splitValue) {
newValue = (axis.axisInfo.splitValue - rawEvent->value)
* axis.scale + axis.offset;
highNewValue = 0.0f;
} else if (rawEvent->value > axis.axisInfo.splitValue) {
newValue = 0.0f;
highNewValue = (rawEvent->value - axis.axisInfo.splitValue)
* axis.highScale + axis.highOffset;
} else {
newValue = 0.0f;
highNewValue = 0.0f;
}
break;
default:
newValue = rawEvent->value * axis.scale + axis.offset;
highNewValue = 0.0f;
break;
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
}
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
axis.newValue = newValue;
axis.highNewValue = highNewValue;
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
}
break;
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
}
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
case EV_SYN:
Refactor key code mapping. Added handling for EV_MSC / MSC_SCAN which typically reports the HID usage associated with a key. This will enable key maps to map keys with HID usages that Linux does not natively recognize. Removed keyCode and flags fields from EventHub RawEvent since they don't necessarily make sense in isolation now that we pay attention to HID usage codes too. Removed the fallback code for mapping keys and axes. In practice, an input device should be self-sufficient. We should not ever need to look at the built-in keyboard's key map. In fact, there usually isn't a built-in keyboard anyhow. This code was originally working around a problem where we weren't loading the key map for touch screens with virtual keys, which has long since been fixed. Change-Id: I0a319bdec44be9514f795526347397e94d53a127
2012-04-11 18:27:33 -07:00
switch (rawEvent->code) {
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
case SYN_REPORT:
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
sync(rawEvent->when, false /*force*/);
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
break;
}
break;
}
}
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
void JoystickInputMapper::sync(nsecs_t when, bool force) {
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
if (!filterAxes(force)) {
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
return;
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
}
int32_t metaState = mContext->getGlobalMetaState();
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
int32_t buttonState = 0;
PointerProperties pointerProperties;
pointerProperties.clear();
pointerProperties.id = 0;
pointerProperties.toolType = AMOTION_EVENT_TOOL_TYPE_UNKNOWN;
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
PointerCoords pointerCoords;
pointerCoords.clear();
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
size_t numAxes = mAxes.size();
for (size_t i = 0; i < numAxes; i++) {
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
const Axis& axis = mAxes.valueAt(i);
pointerCoords.setAxisValue(axis.axisInfo.axis, axis.currentValue);
if (axis.axisInfo.mode == AxisInfo::MODE_SPLIT) {
pointerCoords.setAxisValue(axis.axisInfo.highAxis, axis.highCurrentValue);
}
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
}
Wake screen from external HID peripherals. Added some plumbing to enable the policy to intercept motion events when the screen is off to handle wakeup if needed. Added a basic concept of an external device to limit the scope of the wakeup policy to external devices only. The wakeup policy for internal devices should be based on explicit rules such as policy flags in key layout files. Moved isTouchEvent to native. Ensure the dispatcher sends the right event type to userActivity for non-touch pointer events like HOVER_MOVE and SCROLL. Bug: 3193114 Change-Id: I15dbd48a16810dfaf226ff7ad117d46908ca4f86
2011-03-02 19:23:13 -08:00
// Moving a joystick axis should not wake the devide because joysticks can
// be fairly noisy even when not in use. On the other hand, pushing a gamepad
// button will likely wake the device.
// TODO: Use the input device configuration to control this behavior more finely.
uint32_t policyFlags = 0;
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
NotifyMotionArgs args(when, getDeviceId(), AINPUT_SOURCE_JOYSTICK, policyFlags,
Add initial API for stylus and mouse buttons. Added the concept of pointer properties in a MotionEvent. This is currently used to track the pointer tool type to enable applications to distinguish finger touches from a stylus. Button states are also reported to application as part of touch events. There are no new actions for detecting changes in button states. The application should instead query the button state from the MotionEvent and take appropriate action as needed. A good time to check the button state is on ACTION_DOWN. As a side-effect, applications that do not support multiple buttons will treat primary, secondary and tertiary buttons identically for all touch events. The back button on the mouse is mapped to KEYCODE_BACK and the forward button is mapped to KEYCODE_FORWARD. Added basic plumbing for the secondary mouse button to invoke the context menu, particularly in lists. Added clamp and split methods on MotionEvent to take care of common filtering operations so we don't have them scattered in multiple places across the framework. Bug: 4260011 Change-Id: Ie992b4d4e00c8f2e76b961da0a902145b27f6d83
2011-05-06 18:20:01 -07:00
AMOTION_EVENT_ACTION_MOVE, 0, metaState, buttonState, AMOTION_EVENT_EDGE_FLAG_NONE,
1, &pointerProperties, &pointerCoords, 0, 0, 0);
Refactor input reader to add stylus support. Bug: 5064702 Introduced the concept of an InputListener to further decouple the InputReader from the InputDispatcher. The InputListener exposes just the minimum interface that the InputReader needs to communicate with the outside world. The InputReader passes arguments to the InputListener by reference, which makes it easy to queue them up. Consolidated all of the InputReader locks into one simple global Mutex. The reason this wasn't done before was due to potential re-entrance in outbound calls to the InputDispatcher. To fix this, the InputReader now queues up all of the events it wants to send using a QueuedInputListener, then flushes them outside of the critical section after all of the event processing is finished. Removing all of the InputMapper locks greatly simplifies the implementation. Added tests for new stylus features such as buttons, tool types, and hovering. Added some helpers to BitSet32 to handle common code patterns like finding the first marked bit and clearing it. Fixed a bug in VelocityTracker where the wrong pointer trace could get cleared when handling ACTION_POINTER_DOWN. Oops. Changed PointerCoords so it no longer stores useless zero axis values. Removed editAxisValue because it is not very useful when all zero value axes are absent and therefore cannot be edited in place. Added dispatch of stylus hover events. Added support for distance and tool types. Change-Id: I4cf14d134fcb1db7d10be5f2af7b37deef8f8468
2011-07-27 16:04:54 -07:00
getListener()->notifyMotion(&args);
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
}
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
bool JoystickInputMapper::filterAxes(bool force) {
bool atLeastOneSignificantChange = force;
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
size_t numAxes = mAxes.size();
for (size_t i = 0; i < numAxes; i++) {
Joystick tweaks. (DO NOT MERGE) Ensure that the joystick can always reach -1.0, 0.0 and 1.0 positions even when noise filtering is applied. (Bug: 3514510) Add support for a few more standard axes. Add additional mapping modes for axes. Some axes are inverted from standard interpretation or are actually intended to be split into two distict axes such as left/right trigger controls or accelerator/brake. Add key layout file for a G25 racing wheel and XBox 360 controller to tweak behavior. They work fine without them but the axis mappings are not ideal. Change-Id: I0fddd90309af4dc14d35f34fe99ed6e521c0b7c7
2011-03-04 13:07:49 -08:00
Axis& axis = mAxes.editValueAt(i);
if (force || hasValueChangedSignificantly(axis.filter,
axis.newValue, axis.currentValue, axis.min, axis.max)) {
axis.currentValue = axis.newValue;
atLeastOneSignificantChange = true;
}
if (axis.axisInfo.mode == AxisInfo::MODE_SPLIT) {
if (force || hasValueChangedSignificantly(axis.filter,
axis.highNewValue, axis.highCurrentValue, axis.min, axis.max)) {
axis.highCurrentValue = axis.highNewValue;
atLeastOneSignificantChange = true;
}
}
}
return atLeastOneSignificantChange;
}
bool JoystickInputMapper::hasValueChangedSignificantly(
float filter, float newValue, float currentValue, float min, float max) {
if (newValue != currentValue) {
// Filter out small changes in value unless the value is converging on the axis
// bounds or center point. This is intended to reduce the amount of information
// sent to applications by particularly noisy joysticks (such as PS3).
if (fabs(newValue - currentValue) > filter
|| hasMovedNearerToValueWithinFilteredRange(filter, newValue, currentValue, min)
|| hasMovedNearerToValueWithinFilteredRange(filter, newValue, currentValue, max)
|| hasMovedNearerToValueWithinFilteredRange(filter, newValue, currentValue, 0)) {
return true;
}
}
return false;
}
bool JoystickInputMapper::hasMovedNearerToValueWithinFilteredRange(
float filter, float newValue, float currentValue, float thresholdValue) {
float newDistance = fabs(newValue - thresholdValue);
if (newDistance < filter) {
float oldDistance = fabs(currentValue - thresholdValue);
if (newDistance < oldDistance) {
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
return true;
}
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
}
Add new axes for joysticks and mouse wheels. Added API on InputDevice to query the set of axes available. Added API on KeyEvent and MotionEvent to convert keycodes and axes to symbolic name strings for diagnostic purposes. Added API on KeyEvent to query if a given key code is a gamepad button. Added a new "axis" element to key layout files to specify the mapping between raw absolute axis values and motion axis ids. Expanded the axis bitfield to 64bits to allow for future growth. Modified the Makefile for keyboard prebuilts to run the keymap validation tool during the build. Added layouts for two game controllers. Added default actions for game pad button keys. Added more tests. Fixed a bunch of bugs. Change-Id: I73f9166c3b3c5bcf4970845b58088ad467525525
2011-02-19 01:08:02 -08:00
return false;
Add joystick support to framework. Change-Id: I95374436708752e1a9cff3f85c5b9bc3e0987961
2011-01-15 18:14:15 -08:00
}
Native input dispatch rewrite work in progress. The old dispatch mechanism has been left in place and continues to be used by default for now. To enable native input dispatch, edit the ENABLE_NATIVE_DISPATCH constant in WindowManagerPolicy. Includes part of the new input event NDK API. Some details TBD. To wire up input dispatch, as the ViewRoot adds a window to the window session it receives an InputChannel object as an output argument. The InputChannel encapsulates the file descriptors for a shared memory region and two pipe end-points. The ViewRoot then provides the InputChannel to the InputQueue. Behind the scenes, InputQueue simply attaches handlers to the native PollLoop object that underlies the MessageQueue. This way MessageQueue doesn't need to know anything about input dispatch per-se, it just exposes (in native code) a PollLoop that other components can use to monitor file descriptor state changes. There can be zero or more targets for any given input event. Each input target is specified by its input channel and some parameters including flags, an X/Y coordinate offset, and the dispatch timeout. An input target can request either synchronous dispatch (for foreground apps) or asynchronous dispatch (fire-and-forget for wallpapers and "outside" targets). Currently, finding the appropriate input targets for an event requires a call back into the WindowManagerServer from native code. In the future this will be refactored to avoid most of these callbacks except as required to handle pending focus transitions. End-to-end event dispatch mostly works! To do: event injection, rate limiting, ANRs, testing, optimization, etc. Change-Id: I8c36b2b9e0a2d27392040ecda0f51b636456de25
2010-04-22 18:58:52 -07:00
} // namespace android
Reference in New Issue Copy Permalink