The VisibleForTesting annotations are used by proguard to prevent
stripping code that is needed at runtime for tests. The annotations
were marked with source retention, which worked fine with Jack's
proguard implementation. When we switched back to javac + proguard,
the VisibleForTesting annotations are gone by the time proguard
sees the classes, so the annotated classes and methods can still
be stripped.
Move the VisibleForTesting annotation to class retention, which
matches guava's equivalent annotation.
Bug: 64365384
Test: m -j MtpDocumentsProvider &&
javap -cp out/target/common/obj/APPS/MtpDocumentsProvider_intermediates/classes-proguard.jar com.android.mtp.MtpRoot
Change-Id: I63a467c2f0a21383eb9e4c9563fd6de2d994a373
No longer do we play crazy rename-into-place tricks! Just take
the restored image and treat it as an ordinary new system wallpaper,
and let the system machinery do all its proper work. While we're at
it, we get rid of the now-spurious shenanigans about backing up the
wallpaper under the 'android' package, an overdue and welcome bit
of thinning.
In addition, we now fully migrate a legacy wallpaper image to have
both a source and a display crop in place. We were previously
not generating the equivalent source image, which made N+ backups
unable to store the image properly.
Fix 64269208
Fix 62809439
Test: manual (set wallpaper under M, backup, flash to ToT,
force restore)
Change-Id: If9b26b777f0fda95cf37c3c790a3fa4e82ff0340
Test: set different wallpapers for different users and switch between them.
Test: re-ran cts tests at cts/tests/app/src/android/app/cts/WallpaperManagerTest.java
Change-Id: Ic06d1dc6db26869a2948590863ca9b8ac81c630e
Fixes: 63513694
Test: set different wallpapers for different users and switch between them.
Test: re-ran cts tests at cts/tests/app/src/android/app/cts/WallpaperManagerTest.java
Change-Id: Ic06d1dc6db26869a2948590863ca9b8ac81c630e
Merged-In: Ic06d1dc6db26869a2948590863ca9b8ac81c630e
Fixes: 63513694
Remove convenience method from DocumentsProvider, clients to use MetadataReader directly.
Concentrate mimetype checking in MetadataReader.isSupportedType.
Update FileSystemProvider to use MetadataReader directly.
Test: Updated mtp tests. Other functioanlity manually verified.
Change-Id: Ie1e3d3092b53107f6c980c18b1451290dd2a9653
It didn't lend it self naturally to the multi-metadata return type and was complicating further adoption of the feature in system providers.
Test: manual
Change-Id: I2dde59c7e9609a205614f715c7383b653fa83111
Dumping current themes and scrim states to make debugging easier.
Bug: 63890571
Test: adb bugreport # look for new variables
Change-Id: Ia9bfd1b91d8d2c19930cb896d337ecdfa776e6cb
This effectively reverts 19c4883c85.
The change added a binder death listener on a peer of the AsyncChannel.
Issue: as specified in the bug - this causes a memory leak in apps
using WifiP2pManager (and possibly other clients using AsyncChannel).
The issue is that a reference to the caller is now kept by the death
listener. This means that the caller object cannot be gc'd. This
impacts WifiP2pManager since it didn't have an explicit close() method.
The original reason for the change was to enable clean-up - this can
be done in an alternative mechanism by passing-in a binder from client
to service. Will be enable in a subsequent CL.
Bug: 37443149
Test: all p2p tests, memory usage with test app
Change-Id: Iff044a8cc9c2fc0bc57db4a218b80d0b12424ddd
Isolated UIDs keep coming and going, with their process states
being downgraded. But this shouldn't impact the process state
time tracking in battery stats, since the parent UIDs process state
reflects the overall state already.
This fixes incorrect TOP time reporting in Settings for apps like
Chrome.
Change-Id: I0e94b064cd19f1f2d951e1e88aaba06b5c1c26fe
Fixes: 62162308
Test: Use chrome for a while and observe output of dumpsys batterystats