This patch covers 2 cases. When an app is installed
and the resulting data directory is created for all
existing users. And when a new user is created and
all existing app data directories are created for
the new user.
Change-Id: Iacaba6d9d18d5337e65713960d14efe32006b330
Signed-off-by: rpcraig <rpcraig@tycho.ncsc.mil>
This patch set allows the PMS to parse the
mac_permissions.xml file which contains the
seinfo values. Each package that is installed
on the device will be assigned an seinfo value
based on policy. This seinfo value will help label
the app process and data directory. Modifications
include adjustments to ApplicationInfo.java
to store the seinfo tag per package as well as
adjustments to installd to communicate the seinfo
tag to libselinux.
Change-Id: I61ad1ea12fb6a9a6d0b108ec163bc4bf4c954b58
Signed-off-by: rpcraig <rpcraig@tycho.ncsc.mil>
After discussion, it was determined that duplicate would be less
disruptive and it still fit in the current HAL model.
Change-Id: I2f9cae48d38ec7146511e876450fa39fc92cda55
To support the WiFi service, we need to support migration from the
system UID to the wifi UID. This adds a command to achieve the
migration.
Bug: 8122243
Change-Id: I65f7a91504c1d2a2aac22b9c3051adffd28d66c1
In previous commits, we added the ability to specify which UID we want to
target on certain operations. This commit adds the ability to reach those
binder calls from the KeyStore class.
Also fix a problem where saw() was not reading all the values returned via
the Binder call. This changes the semantics to return a null instead of
failing silently when it's not possible to search.
Change-Id: I32098dc0eb42e09ace89f6b7455766842a72e9f4
The HeartbeatHandler for the System Server Watchdog has been running
on the wrong thread due to a race condition in initialization. It's
designed to run on ServerThread, so that it can catch lockups in the
main looper of the System Server. It has been running on
ActivityManagerThread instead, so it does not detect lockups on the
ServerThread as it should.
ActivityManagerService is calling Watchdog.getInstance() before
ServerThread calls Watchdog.getInstance().init(), so the handler is
being bound to the ActivityManagerThread instead of the ServerThread.
Explicitly bind HeartbeatHandler to ServerThread, so that the Watchdog
catches lockups on this critical thread.
Change-Id: Iccb184ac3adb817feb86ed4ee0e50e443bf74636
The root cause is:
There is a defect in window manager service: When a new
activity that can be ime target is added into window manager
but the Z order of input method window don't need to be
changed, then the target app token of input method window
would not be updated to new one. This defect may cause that
the layer of input method window is calculated incorrectly.
The solution:
Correct the target app token for input method window.
Change-Id: I008311e3c9b1cf5fc320b614d8675c183c506d50
Different kernels seem to handle ADDR_COMPAT_LAYOUT differently,
sometimes passing it to its children, sometimes not. If it's not
passed to its child successfully, we can end up in a restart loop.
Instead of testing for the presence of ADDR_COMPAT_LAYOUT, use an
environment variable instead, which is handled more predictably.
Bug: 8392487
Change-Id: Ia531dd2abb4e1cd46f3430d844e644f53581f530
For the emulator, we want people to see memory as it
actually is, not how we're hacking around buggy apps. Don't
set ADDR_COMPAT_LAYOUT on the emulator.
For reasons that I don't understand, personality(ADDR_COMPAT_LAYOUT)
does not persist across an exec on the emulator. app_main gets
into a tight loop restarting itself because of this. This change
also works around that bug.
Change-Id: Ia73a7d2d623c25cf39d248145d97307945d554da
"[+>" more icon was never show in status bar because
the member variable for this icon was not initialized
from resources. This fix enables "[+>" icon to appear
in status bar when the number of indications in status
bar becomes large.
Bug: 8368569
Change-Id: Ieb3412eed831052d69c0cf63c9b4230c38171e4a
This is to accept both the "transparent" and "opaque" ECC private
keys. "Transparent" keys provide structured access to their key
material -- these are instances of ECPrivateKey. "Opaque" private
keys are not required to provide structured (or even any) access to
their key material -- these are instances of PrivateKey.
Change-Id: Ib22e18b45b638b429f994ed965416c753226c4ee
When copying a link from a bookmark and then pasting it into
a textfield a NullPointerException will occur.
A ClipData.Item is not guaranteed to always contain a text string
and therefore getText() can be set to null.
Using method coerceToText() instead of getText() makes sure that
a text string is always returned.
Change-Id: I81343c0371835a3a7a52045dcd1760e69e59a967
The DatePickerDialog is using DateUtils to format the dialog's title
and the DateUtils class does not work with dates outside to the
specified range.
For example, if user selects 2038-03-07, DatePickerDialog shows
1902-01-30 on Title.
The reason for the DateUtils class not being able to format dates
outside the range of 1902 and 2036 is because internally it is using
the Time class which does not support such dates.
To fix it, use Calendar class in DataUilts format method.
Steps to reproduce:
1. Settings -> Date & time
2. uncheck Automatic date & time
3. Set Date
4. choose any date before 1902 or after 2037
5. update wrong date on Title
Bug: https://code.google.com/p/android/issues/detail?id=13050
Change-Id: I003266765751b5c340426af84daef271f39f771e
Remove some preloaded-classes and fix typo for deleting log messages
as the followings during booting.
W/Zygote ( 163): Class not found for preloading: android.bluetooth.
BluetoothAudioGateway
W/Zygote ( 163): Class not found for preloading: android.bluetooth.
HeadsetBase
W/Zygote ( 163): Class not found for preloading: undroid.content.Ab
stractThreadedSyncAdapter$SyncThread
Change-Id: Id02e0cccaeb150f48e12d67b2d691df828842e4c
Add support for the 'c' format character, required in many non-English locales.
Reimplement 'c' and 'E', and 'L' and 'M', so they correctly interpret 5-count
pattern characters.
Replace the old incorrect class documentation with a pointer to the
well-maintained libcore equivalent and the Unicode UTS to which these two
implementations are supposed to conform.
Deprecate the useless constants for pattern characters. No one sane is going
to write MONTH + MONTH + MONTH + MONTH instead of "MMMM".
Correct the documentation for getLongDateFormat and getMediumDateFormat.
Also fix DateUtils.getStandaloneMonthString for LENGTH_SHORTEST.
Tests are in https://android-review.googlesource.com/53291.
Change-Id: I4dda8b18070f05ccdc11c1f0a9381a9d233db4e8
Natural languages differ so much and in such odd ways that you can't
use getQuantityString as an "if" statement. It's really just for grammaticality.
This is explained well in
http://developer.android.com/guide/topics/resources/string-resource.html#Plurals
but we need to make more of an effort to motivate people to read that, and to
at least get the most important point across if they don't read it.
Change-Id: I549b9f3563462c45f2dea34c558185e0714127cd
The bug we're fixing here is that languages that don't
distinguish the "one" case grammatically (such as Japanese)
would say the equivalent of "In 1 day" rather than "Tomorrow"
because of the misuse of getQuantityString.
This has the side-effect of switching us over to the CLDR
strings for relative day names, which have consistent capitalization;
the Android donottranslate-cldr.xml strings varied even within
a language, so although this is a change, it seems like a step
in the right direction.
In a future change, we should actually push all relative
day formatting down into icu4c.
Bug: 7098707
Change-Id: Ia2f9af3d18c441d6093dd5da7956a3d0130e5b06
