| Auto-Submit | +1 |
Hello, please try to look on this bugfix. It's a small change in Android part. Please also start Dry run for me (I'm new here). Thanks a lot
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
Peter, can I ask you to take a look? This involves the web app splash screen.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
This fix mitigates the symptom in the CCT back-press path, but the underlying corrupted splash state is still reachable through other exits (system kill, config change, notification return). The real bug appears to be in SplashController#bringSplashBackToFront() not guarding on mWasSplashHideAnimationStarted, and in TwaFinishHandler reporting success before the async WebAPK IPC completes.
Could you either (a) fix it there, or (b) leave a TODO(crbug/40746880) and file a follow-up so this doesn't get forgotten?
Bug: 40746880put `Bug: 40746880` at the bottom, below `Change-Id`
if (isScreenPinned()) return true;there is cases where this path hit without being a PWA/TWA/WebAPK like Auth Tab, Payment Handler, are we sure this is not too broad?
can we add tests for it?
`ShadowActivityManager.setLockTaskModeState(...)` in covers this cleanly; add cases for LOCK_TASK_MODE_NONE, LOCK_TASK_MODE_PINNED, and LOCK_TASK_MODE_LOCKED against TWA + plain CCT intent providers.
optional nit: wonder if these could be simplified and condensed?
it kinda duplicates the isScreenPinned documentation.
return am != null && am.getLockTaskModeState() == ActivityManager.LOCK_TASK_MODE_PINNED;getLockTaskModeState() also returns LOCK_TASK_MODE_LOCKED for kiosk mode, in which finish() is blocked identically. Either compare against != LOCK_TASK_MODE_NONE (and rename isScreenPinned() -> isInLockTaskMode()), or justify pinned-only in the Javadoc.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |