Bug: 188081026
Test: no op change, should compile
Test: ReorderWidgets
Change-Id: I20367974e5a4cead406e18eb66dafd4d59651b2a
Merged-In: I20367974e5a4cead406e18eb66dafd4d59651b2a
Moving a few calls into the UI helper executor thread, as per the associated bug. duplicated the lines so the synchronous nature of clearing previous state wouldn't affect potential timings.
Test: local testing, rotating screen. quick switching apps.
Bug: 257470365
Change-Id: I59abad9ba51c0b19d467646752aa34ae7ef3ec1d
If freeform tasks are shown, skip applying depth effect and scrim
changes. These cause the background to flicker while showing freeform
tasks and for example showing the transient taskbar.
Bug: 263264985
Test: swipe up to show transient taskbar, observe launcher background
does not blur or flicker to black
Change-Id: I5b10d0f0c7065e903cb761488367c02d7e31d8b2
This fixes a concurrency issue where HingeSensorAngleProvider was being stopped and started at the same time in a thread-pool after a fast fold/unfold, despite not providing concurrency guarantees.
In sysui, the background executor provided was already single threaded, so no issue arisen. From Launcher, THREAD_POOL_EXECUTOR was provided.
In a follow up cl, I'll add a @SingleThreadBackground annotation to the executor used in the unfold lib.
Bug: 261320823
Test: manually stress tested fold/unfold.
Change-Id: Iccf1f1f7246d8592d4d80a032479aa75f0050655
Bug: 263913711
Test: Manual, i.e. testing predictive back for four scenarios: Back to all apps, Back to home. Both with the closed app icon visible or not.
Change-Id: Id87686098e6ad413819dc75545071da00291db65
* Check was originally added with rationale in
b/204384193#comment26, however presently we
do provide insets to SUW as of b/253538830
so looks like now we are getting the correct
info from SysUI
* There is a possibly related issue where buttons
are not tinted correctly on the final page of
SUW (b/265238238)
Test: Went through setup manually via adb
and observed that on light theme with dark
dialog scrim the button changes color
adb shell am start -a android.intent.action.MAIN -n com.google.android.setupwizard/.SetupWizardTestActivity
Bug: 256521774
Change-Id: I080033e21d22fbc366b7248b2d647fec610a5fb9
Starting from target SDK U, we will block creation of mutable
PendingIntents with implicit Intents because attackers can mutate the
Intent object within and launch altered behavior on behalf of victim
apps. For more details on the vulnerability, see go/pendingintent-rca.
From a quick analysis, we concluded that the PendingIntent here was only
destined to the test app/to the app, so it was made explicit. Reviewers,
please call out if this is not the case.
Bug: 236704164
Test: atest NexusLauncherTests:com.android.quickstep.DigitalWellBeingToastTest
Test: atest RequestPinItemTest
Change-Id: I81311b0588b7b712c97de6cb016a62bb91c7a77c
into its own class.
Bug: 251502424
Test: Compilation threw no errors and user unlock behavior worked
correctly.
Change-Id: Ifa42dc32f90dfa4fda8df0e52811ddfe20cc5a9b
This change makes it so that a smoother animation will play when the user cancels the splitscreen flow by tapping the staged app to launch the app in fullscreen. Since this was the only blocker to activating the feature, this change also enables launch-from-staged by default and removes the flag.
Fixes: 257513449
Test: Manual
Change-Id: Iabff2fc72bfcd6975d216b5cd86e0df7a772b308
While animating DesktopTaskView to/from fullscreen, hide the background.
Background should only show up when this taskview is shown in recents.
While it is animating, we don't need to show it.
Bug: 263264985
Test: manual, trigger transient taskbar, observe that background does
not flicker
Change-Id: I76d0eb1857645b7f04a0b2e1eebe2393c252c257
This uses ktfmt on the PREUPLOAD hook to check file formatting.
Bug: 264851618
Test: manual
Change-Id: I8bdeabd9c1ed571a464707e6912419ad3190a169
Merged-In: I8bdeabd9c1ed571a464707e6912419ad3190a169
* Translation gets scaled along with the parent,
so the secondary thumbnailView which has a translate on
it gets exaggerated as the parent goes to fullscreen
scale.
* Now we inversely scale translationX/Y properties,
however it seems like the interpolation is still slighly
off at the beginning of the animation (or maybe some other
value, such as pivot, etc)
Bug: 260814999
Test: Manually slowed down animation and observed
Change-Id: I831476b4ffb582fecbd166525ceea33623e66bdd
Merged-In: I831476b4ffb582fecbd166525ceea33623e66bdd
