This CL only implement the handwriting initiation for milestone 1:
a) handwriting is only initiated for a focused editor.
b) user must start handwriting within the editor's boundary.
Bug: 209854913
Test: atest FrameworksCoreTests:HandwritingInitiatorTest
Change-Id: Iab6d83f58aa74f0af8091731c80bbdb2eb29c024
BatteryUsageStats is created under a BatteryStatsImpl lock. One of
the elements of BatteryUsageStats is the battery history buffer Parcel.
Once the BatteryUsageStats object is created, the BatteryStatsImpl lock
is released and the history buffer parcel continues to be appended
by BatteryStatsImpl. The Parcel may even be reset altogether if the
battery stats session is reset. The BatteryUsageStats object is parceled
during the getBatteryUsageStats binder call. Any modification of the
history buffer concurrent with parceling causes a crash.
Bug: 194256984
Test: atest FrameworksCoreTests:BatteryUsageStatsTest FrameworksCoreTests:BatteryUsageStatsProviderTest
Change-Id: Ifb03a32275dfbea172cd28309a42349d6dd4bcd5
When work profile disable by quick settings, the
user would be locked then we cannot query installed
input method lists for work user by existings API
InputMethodManager#getInputMethodListAsUser().
Adding a new API to query input method services
regardless of the user state. This API currently
only used by Settings, it shouldn't affect the
original behavior for other usages.
Bug: 210083408
Test: verify we can get input methods list after
disable work apps by quick settings
Test: atest CtsInputMethodTestCases
Change-Id: I54d5dbec7e76d6a68935007ed3af0641f717a7c5
Add a extra variable to store whether the local traffic is
exempted from the VPN.
Bug: 184750836
Test: atest FrameworksNetTests
Change-Id: I70301901c2374ded9fdf179d0ce64e7bb2d7b1ab
1. Assert the logging package name matches the one we read off of
instrumentation at runtime, instead of hard-coding a package that will
only match in one version of the tests.
2. Set up infrastructure for UnbundledChooserActivityTest to disable
part of the APS availability test that doesn't make sense in the
unbundled version.
3. Expose the name of an extra that we now need to use more widely.
4. Disable a test that seems non-trivial to fix for the unbundled
implementation. This could reflect a real issue with the unbundled
behavior and should be addressed ASAP, but I don't want this to block
ag/16516860 so we can at least get some baseline tests running.
Test: `atest ChooserActivityTest`. Also tested the unbundled version
(slightly modified from ag/16516860 based on the changes above).
Bug: 202166045
Change-Id: Icfca680aa437a9860548d46f9e221b19df36b5a5
This is important in cases where we want to add/clear overrides for
multiple apps at the same time, without having to make a binder call for
each package and have the compat config save the changes to file and
invalidate the cache each time.
Bug: 199730202
Test: atest FrameworksServicesTests:CompatConfigTest
Test: atest FrameworksServicesTests:PlatformCompatTest
Change-Id: I3b4a3d9db12fdf99eb78c486bdd8e248fff92e50
This change uses the current small location icon. This change is flagged
and off by default.
Bug: 191503437
Test: atest SystemUITests
Test: manual
Change-Id: I98195b2c607aeec36d56af75ab20a056b4d44e2f
Test: Make sure apps with custom animations set using WindowAnimationStyle actually use the custom animation when shell is enabled
Bug: 206960607
Change-Id: I1d982870837d67c74f76dd0a90aa5260d9441aba
Since NetworkStats related code will be moved to the module.
multiplySafeByRational in the NetworkUtilsInternal cannot be
accessed after that. Thus, create another utils class that
will be moved with NetworkStats code and put the function
into it.
Test: atest NetworkStaticLibTests:com.android.net.moduletests.util.NetworkStatsUtilsTest
Bug: 204830222
Change-Id: I96f3ac02e57b7325ed53988285770f478dee529e
Make the transport AIDL async and introduce a callback AIDL for passing
back results. A follow-up CL will implement the client-side portion of
the callback class and update BackupTransportClient to use it.
Bug: 202716271
Test: m -j
Change-Id: I79cb376bd2a805bf2739470a3f538207ae6bd9a0
This is a safeguard against cases like Bug 203086369, where the UI
thread can be locked up if IC#endBatchEdit() never returns false no
matter how many times it gets called.
With this CL, the system will have a max retry count on how many times
it attempts to call IC#endBatchEdit(), which is currently set to 16,
then fall back to InputMethodManager#restartInput(View) if it hits
such a limit.
This CL also introduces a volatile type cache to early fall back to
the fallback implementation if the same type of InputConnection
implementation class appears again.
As long as InputConnection#{begin,end}BatchEdit() is correctly
implemented, there should be no developer observable behavior change
in this CL.
See the corresponding test CL [1] about when the fallback
implementation is used.
[1]: Ifb80015ab0f0c32c917a80c4e5f60e836648e7b4
Bug: 203086369
Bug: 208941904
Bug: 209008342
Test: atest CtsInputMethodTestCases:InputMethodStartInputLifecycleTest
Change-Id: I109e0c26d8249fc2e01323e3e1cb36395fa7cc97
Revert "Suppress NewApi warnings for @SystemApi -> public APIs"
Revert "Add VpnServiceBuilderShim for VpnService.Builder"
Revert submission 1551943-vpn-impl
Reason for revert: <DroidMonitor-triggered revert due to breakage https://android-build.googleplex.com/builds/quarterdeck?branch=aosp-master&target=mainline_modules_x86_64-userdebug&lkgb=8007224&lkbb=8008168&fkbb=8007902 >, bug b/210979001
Reverted Changes:
I0e7aa077a:Add VpnServiceBuilderShim for VpnService.Builder
Ib12f5ab39:Suppress NewApi warnings for @SystemApi -> public ...
I59b9185cf:Unhide RouteInfo#getType and related fields
Ie5b62b2b2:Unhide IpPrefix(InetAddress, int)
I993a32d40:Add CTS tests for exclude VPN routes APIs
Ib24b2d3fb:Suppress NewApi warnings for @SystemApi -> public ...
Ic3b10464a:Add APIs that allow to exclude routes from VPN
Change-Id: I89209357ab4bb206d206adcf9fd6d3498ce6ef4c
BUG: 210979001
