Move padding around so that the shadow can blead into the padding.
Bug: 37754078
Test: Pressed button, took screenshot and checked for clipping
Change-Id: Iaed2c7e7c432061878af410ee270de97764bcadf
We were processing escaped strings too early, before
parsing of values into types. Now the escaped strings get
processed when they are being flattened.
Bug: 37715376
Test: make aapt2_tests
Change-Id: Ic59aa2e3a20c40756c219752ff74b2a4f8a602ba
Package IDs greater than 0x7f are interpreted as negative
numbers in Java's signed integer. The proper check for existence of
a resource ID is that it is not 0.
Bug: 37498913
Test: none
Change-Id: I446fb6abb514bf7cf2d0dcbfbd81dd5718cd2cb4
The entire View code base checks IDs against View.NO_ID except
findViewById(), which checks to see if the ID is negative.
Any package ID > 0x7f is interpreted as a negative number in Java
(no unsigned ints), so this check prevents the use of IDs > 0x7f.
findViewById is final, so support library workarounds are not possible.
Instead, IDs (@id/foo) are just sentinels, their values don't matter.
If building for pre-O devices, rewrite any references to these IDs of
the for 0xPPTTEEEE, where PP > 7f, to 0x7fPPEEEE.
The symbol table will check for potential collisions against the base
APK, so this should be safe.
Bug: 37498913
Test: manual
Change-Id: Ife3bbd29db287757ef8a2ffd83053d97f1db2613
This fixes the glitch of having transitions be calculated incorrectly
when starting new activity when other windows still exist. For example
launcher3 has minus one window exist even if the main activity is
destroyed causing incorrect transition going home. This change only
allows the calculation to occur when the app window is not relaunching.
Test: manual - launch app, rotate screen, go home
Change-Id: I7e323486ee6e05ba4c8704832ca0f9d6e648ac90
Fixes: 36113180
This is a temporary solution, in the future we will investigate adding
an API that requires the app adding a setting to specify if the setting
should be visible to Instant Apps or not.
Test: run instant app from b/37765840, see denials are gone.
Bug: 37765840
Change-Id: Ib256d375d2a07f743b360c61ed25c9857aafa36c
If Bluetooth service calls onBluetoothA2dpDeviceConfigChange() before
setBluetoothA2dpDeviceConnectionState(), there is a chance that the
first method fails leaving the A2DP device state disconnected in audio
policy manager while it appears as connected in AudioService causing a
failure to send the AUDIO_BECOMING_NOISY intent.
Bug: 37687852
Test: Verify AUDIO_BECOMING_NOISY intent is sent even if race
condition occurs.
Change-Id: I014d145e7da5e7d267991ffb2ff50626e71247eb
When a local variable (cachedThumbnailData) gets out of scope,
the interpreter might still keep a reference to it in the
register. Thus, we extract the handling of this local variable
into a method such that it gets removed from the register.
Test: Open recents, go home, take dump, make sure no reference
from HandlerThread to ThumbnailData is still existing.
Fixes: 37630958
Change-Id: If25456b212341be7a66882a0adfd5afb4a64ce23
The abbreviation is not in common use. Also remove FBE from
documentation as it also isn't used elsewhere.
Test: Build success
Bug: 37621349
Change-Id: Icf19be5e96e71dcd45aa7cac8f58b05b6d77d02b
Rather than relying on stderr being wired correctly.
Test: Tests for JNI integration in studio-master-dev.
Change-Id: I1c7e7130bc4e56d61de259a9596bee3a6b3520fb
