Doesn't cover home-long-press or overview, but those appear to be
covered by other events.
Test: runtest systemui
Merged-In: I590b52765aaef3c2bf573ce8ab980aa992b1d5b0
Change-Id: I630356245e223a3fb8d4d2fe11be5d498227c0e3
Fixes: 25931698
Because English time patterns use uppercase letters by default (and a
comma to separate date and time when both are represented), we were
concluding they need internationalized input. Although they
literally do, let's keep the world simple and assume they don't need
internationalized input.
Compared to Nougat, we will now accept uppercase AM and PM and comma
for English if the IME allows them, we just continue to not signal
that an internationalized layout is needed.
Test: CTS tests pass
Bug: https://code.google.com/p/android/issues/detail?id=2626
Bug: https://code.google.com/p/android/issues/detail?id=82993
Bug: 8319249
Bug: 33276673
Bug: 34394455
Bug: 37079150
Change-Id: I82bfde323ba49ae1a27ff5c2e729063b7d81dcc8
The injected tiles code currently treats the string metadata value
for the title key in the Activity's manifest as the title itself.
When the language on the device is changed the title remains in the
previous language until a reboot.
This CL will make injected tiles attempt to treat the string metadata
value as a resource variable to get the integer res id and then
properly load the localized title using the res id. To be backwards
compatible, if the res id is 0, we treat the string metadata value as
the title itself (current behavior).
Bug:36470694
Test: make RunSettingsLibRoboTests
Change-Id: I3963534a3a3a27d0455c4ed1ad4fa574b03d67fc
If we are measuring storage and a volume is removed at the same time,
we can enter a state where we previously verified the volume existed,
but it no longer does. This causes an NPE.
By adding in a null check, we can avoid this crash.
Change-Id: Ib8dbf05102a122bdf4bb6063374e993a1de68425
Fixes: 36689190
Test: None
The (void*)buffer.get on ARM32 is 4 byte, so the calling convention
will put the argument in [sp, #12]. However, the caller actually
expects a long (the signature of gGraphicBufferClassInfo.builder),
which means it will expect it to be in [sp, #16]
Test: Tested on mtk device
Fixes: 36631082
Fixes: 36974487
Change-Id: I0f723125e612d096c0d76ca3360d895f3f23f286
(cherry picked from commit 98dd5d9a85e8911cf41dea6198d4111f737a5892)
android.display being in the foreground cpuset group is an issue. As
seen on M/S, during heavily CPU load it is not given core 3 even though
it might be free and causes jank. This patch adds the thread to the
top-app group to ensure it is placed on all cores during scheduling
decisions.
Doing this required a couple of changes:
- new API to set per-thread cpusets
- changes to DisplayManagerService to set the thread to top-app group
- changes to SystemServer to set the policy toward the end, as doing it
during start of the DisplayManagerService was in issue (issue being
SystemServer calls setSystemProcess.. -> setProcessGroup which overrides
the group settings for threads in the system server process, including
android.display)
Bug: 36631902
Test: Boot and make sure android.display thread is in the top-app group
Change-Id: Icc394ea0ffcf159d11728ad38de114234a29d20f
Signed-off-by: Joel Fernandes <joelaf@google.com>
(cherry picked from commit 474d311cb098e86c078c3f615e1161e2854f1847)
The previous charging string is "75% - Charging", this cl add lower
case string to make it "75% - charging".
Bug: 36558216
Test: Build
Change-Id: I683843de0b6cbd45bea6a465c7b6d8ee0b7c4f93
This toggle is used to decide whether to show charging method.
If it is off, then "Charging over ac" will become "Charging"
Bug: 36558216
Test: RunSettingsLibRoboTests
Change-Id: I98b1f7e7b77d7ce0a7e1aae29e55a33ac79ff0d7
(cherry picked from commit b3c33f63ed120cf4758ec67426bb45291e289fce)
FingerprintService maintains a mCurrentUserId field which keeps track of
the active group/user id in the underlying daemon. This field, however,
can become invalid after the daemon crashes and before
FingerprintService reconnects to a new daemon. (mCurrentUserId is only
set upon updateActiveGroup().)
Previously, mCurrentUserId is wrongly used in the logic of
isCurrentUserOrProfile(), resulting in false negatives when
mCurrentUserId is invalid. In this CL, we use
ActivityManager.getCurrentUser() instead, which has no dependency on the
state of the daemon.
This will also reduce the complexity around mCurrentUserId, and
hopefully will make future clean-ups easier.
Test: manually kill daemon and verify fingerprint functions normally
Bug: 36152864
Change-Id: I4250fad0e7c88bde95a1da1c89e71b4931ee3b8a
Previously, phone number fields defaulted to locale direction, which
created problems for apps that were not aware of LTR requirements for
phone numbers.
Now, we look at the direction of the digits for the locale, and use
that to determine the direction of the edit field. (For practically
all major RTL locales, that direction is LTR.)
Test: Manual
Bug: 33643035
Change-Id: I17c70d8462bd403ea6866057971105f1f5772ba3
... by extending the RecommendationInfo to store the inet addresses not
just the count. Thankfully the recommendation plugins already contained
all the printers, not only the count.
Also expose the print service state as system APIs as they will be used
by the default print service.
Bug: 35765644
Test: Ran AddPrinterActivity on network that had printers to discover
Change-Id: I929a177adb75f6848848ba30a472cf0343d2e67e
This path changes a dangerous lock path in reportNetworkConnectivity().
This methods is called outside of the main ConnectivityService handler
and takes a lock on a specific NetworkAgentInfo whose connectivity
status is being reported.
While this lock is held, reportNetworkConnectivity() goes on and query
the network policy state for that network, which may ends into
NetworkPolicyManagerService.
Instead, the lock on NetworkAgentInfo is only held long enough to make a
copy of LinkProperties, which is then passed to
NetworkPolicyManagerService without that lock.
Bug: 36902662
Test: could not repro b/36902662, reportNetworkConnectivity() works.
$ runtest frameworks-net
Change-Id: Iac4b75bcecbdddb0ac695c8b1a87ae755f62f47f
1) Switch AccessibilityButtonChooserActivity to non-exported
2) Update AccessibilityButtonChooserActivity's intent action to
be CTS-compliant
3) Switches to correct permission to validate a click of the
accessibility button originated from the status bar service
Bug: 36866328
Bug: 36976440
Test: Manual / Created app to validate launch protection
Change-Id: Ifcd5e8dbbe683dc99bb900a5870a9ba374fbd5d4
Test team is still relying on this. We'll keep this until the virtual display input code goes in this week.
This reverts commit abcc8e54f636bd4eaff5f85d54d5ff8b75508bb9.
Change-Id: Ic591f813289191b09367a76be3221183562256a7
