With accessibility services on (e.g. Talkback), some MotionEvents get
transformed into HoverEvents. Notably in the case of TalkBack, tapping
on PIP will generate a pair of Hover enter/exit event, which then causes
the menu to immediately show/hide, and the user is unable to do
anything.
We will disable the hover behavior for Accessibility services, and allow
TalkBack users and such to open the menu via other menus (just tapping
it twice so it shows menu similiar to a regular user flow).
Bug: 152588263
Test: Enable talkback, enter PIP, tap once and then again - menu shows
Change-Id: I28c8faa0032bc2c7b8fa1ef5a9efde94a590293e
The number of spacers should be how many tiles we can fit, minus
how many we have.
Bug: 152628601
Test: manual
Change-Id: I7cddaba3282068ab6f059baff5ad774ae9bf8b11
Make the iorapd compilation more robust by introducing retries and
switching to direct iorapd dumpsys commands instead of indirect
commands.
Test: forrest runs
Test: am instrument (manual)
Bug: 152322429
Change-Id: I7f75f0df95a1a5272237295a75bed8c34f95f859
The headertext wasn't animating to the title anymore because
the order of the low-priority call has changed in a refactor
We're now looking at the datasource directly instead of looking
at the row state avoiding the ordering issue
Fixes: 151253867
Test: add min priority notification, observe header transforms to title
Change-Id: I7dc8f5fdef4f1f449064e66bcc63a85f7e69a185
Using service connector - as would DataLoaders do - instead of calling
service method directly.
Bug: b/153530451
Test: atest IncrementalServiceTest
Change-Id: I43b9ffa235204b073465c19ab32c0435d2dfe755
Abstract test utilities into two new functions: writeAttribution and
parseStatsEventToLogEvent
Bug: 149590301
Test: bit statsd_test:*
&& bit statsd_benchmark:*
Change-Id: I5f77646b6d2d828344b8b6de8777a60d98f96d58
Items:
- use proper style
- align left and right sides of seek bar with other elements
- still show bar when progress is disabled
Bug: 150724977
Test: manual - play music and look at seek bar
Test: manual - play iHeartRadio for example with progress disabled
Test: manual - play spotify which has seek disabled and check style
Change-Id: I340df584ae4dfa0331eedd0e9fd20fcec19ba8da
To avoid misunderstanding while checking the log.
Bug: 143573504
Test: CtsAppTestCases:ActivityManagerTest#testAppNotResponding
Change-Id: I162aab785cd83db5db6cb33279f0b4ab0a458e94
As the final enabler for the BLASTSyncEngine, we add a new relayout
result BLAST_SYNC. If the WindowState is participating in a BLAST sync
during relayout we return it to the client. If we return it to the
client the client will direct it's next draw in to a BLAST transaction
and report that transaction via finishDrawing. You can now observe
the BLASTSyncEngine working end to end in TaskOrganizerMultiWindowTests.
We do a few small clean-ups in TaskOrganizerMultiWindowTest while we
are there.
Bug: 153561718
Test: TaskOrganizerMultiWindowTests
Change-Id: I719b731350b942aafa444a33972aaef8973422ea
If a non-resizable activity can fill the non-decor space,
its bounds should not exclude decor bounds because the
activity and decor can show as a continuous area. otherwise
unexpected letterbox may be created and IME cannot attach
to it due to the activity doesn't match parent's bounds.
Fix: 152197007
Test: SizeCompatTests#testAspectRatioMatchParentBoundsAndImeAttachable
Change-Id: I61d2c99702fb8642a3bf5728949489b6b98eb508
Bug: 153104923
Test: m -j CarSystemUI && adb sync && adb shell pkill systemui
Manually tested:
1) Boot with no security;
2) Boot with security and unlock;
3) Boot with security, cancel unlock and switch to user with
security;
4) Boot with security, cancel unlock and switch to user without
security.
Change-Id: Id8f13db3d5b53ffd6b3d8624ea2e3f51982b4910
See go/br-fr-live-wallpaper for context
Bug: 30205325
Test: 1. Set live wallpaper (from pre-installed / from a 3p app Muzei)
2. Run Backup Now
3. Wipe the device
4. Restore (regular/deferred)
5. Verify the same live wallpaper is set
Change-Id: Iacdb4e4b4bd895740f0a7f72f075a24fae1ae368
Previously, disabling the consuming logic was
done based on whether we target R.
But this:
- didn't work for Q apps that already use the new API
- causes a bit of unnecessary behavior churn
- requires special-casing the IME
Instead, we look at whether the app has requested to
handle insets with the R setDecorFitsSystemWindows API.
Fixes: 152384150
Test: atest DecorInsetSdk29Tests DecorInsetTests
Change-Id: I754851d0c11dccd21c2fa011cfdc3f0d7396cf60
StatsPullerManager should be declared on the heap because wp's can only
point to objects on the heap.
Test: bit statsd_test:*
Test: atest GtsStatsdHostTestCases
Test: atest CtsStatsdHostTestCases
Bug: 153237308
Change-Id: I579375f5bd2db9557f108f39916d69f368865478
This will scale better as we add more views and gestures.
Test: manual (fixing test suite now, just getting out for review!)
Bug: 152877243
Merged-In: I46001c24f4de58c3d70b7629a22b930fa791b77a
Change-Id: I46001c24f4de58c3d70b7629a22b930fa791b77a
We need that to enable CTS for Environment#isExternalStorageManager API
Test: build
Bug: 152478019
Change-Id: I8c24c08991558629c83a25b4bed89c817723d732
module_current is always built from sources, whereas the
other SDK jars are being transitioned to be composed of
a set of stubs. This avoids the dependency cycle of module
stubs depending on themselves.
Bug: 144149403
Test: m
Change-Id: I31c6d0fa958aa502131352716a9f950828b7829c
The default specifies sdk_version correctly, so use
the inherited value to make it easier to make global
changes.
Bug: 144149403
Test: m
Change-Id: I3e9e3cca6fe0b6dfc7645b91f5cab645350c0f9a
The default specifies sdk_version correctly, so use
the inherited value to make it easier to make global
changes.
Bug: 144149403
Test: m
Change-Id: Ie78e2a934ed52cc9ad31dc6a3c0fb7e0c4b915a1
The default specifies sdk_version correctly, so use
the inherited value to make it easier to make global
changes.
Bug: 144149403
Test: m
Change-Id: Ibc2a1af60f10dd5d270e502011a70e20e7ed934a
The root causes of this issue are:
1. WindowTokenClient#attachContext makes WindowTokenClient has strong
reference to WindowContext, which leads to WindowContext cannot be
GC'd.
2. WMS#removeWindowToken needs MANAGE_APP_TOKEN permission which
normal apps don't hold.
This CL does following things:
1. Use weak reference instead on WindowTokenClient#mContext.
2. Relax WMS#removeWindowToken to check callingUid if
MANAGE_WINDOW_TOKEN permission is not held
3. Deliver config changes to the client side in
WMS#addWindowTokenWithOption
4. Some minor fixes
fixes: 150812449
Bug: 150715095
Test: atest WindowContextTest
Test: atest WindowManagerServiceTests
Test: atest WindowTokenTests
Test: atest WindowManagerPermissionTests#testMANAGE_APP_TOKENS
Test: atest
WindowManagerPermissionTests#testADD_WINDOW_TOKEN_WITH_OPTIONS
Change-Id: I9f1d73af2abb78fc9844e6d9eb25e9f0293514e7
Fix registering PiP transition callback in PipManager's constructor two
times, and thus receiving two calls every time transition state changes,
which caused the linked below bug.
Bug: 153517179
Test: manually
Change-Id: Ib948b4911f7c6b8c3b0abecb3b293c2e27aefda1
