The Chrome team was planning to provide a new prebuilt SDK last week
which returned to targeting the official R SDK level, but other
challenges prevented them from doing so.
They still intend to land an updated prebuilt which targets R, but
to unblock testing this change makes that change on their behalf;
when we see a very specific Chrome version code we force the target
SDK back to R. This code will safely become a no-op during their
next prebuilt, which should have a different version code.
Bug: 183905675
Test: manual
Change-Id: I211e7550f489294f90d8729b518ec9a70a37ec96
Change get -> valueAt and get the actual base.apk ApkAssets for the
package. Also fixes overlay reference visibilty, which was broken with
the AppsFilter cache.
Bug: 184834206
Test: atest OverlayReferenceMapperTests
Test: atest OverlayActorVisibilityTest
Test: atest AppsFilterTest
Change-Id: I01e6a39b36a0e03c2cb9d48e705fe02c1a63c169
Forgot the docs update part of the feedback. Also enforces the param
non-nullablility.
Bug: 184891031
Test: atest DomainVerificationManagerApiTest
Change-Id: I4c90f69bc38c9760a76555ee8c8ad121cac997b6
We use a fake name to prevent feature leak, we should change it back
before API freeze.
Bug: 181179744
Test: build pass and can run service correctly
Test: atest CtsTranslationTestCases
Change-Id: I9c5752e4ce37ff4af8175b6d7527b8a8117588d3
Some testing devices were flashed to a build which contains a previous
version of AppSearchShortcutInfo, the schema wasn't in use and should be
updated to the latest version directly.
Bug: 185186093
Test: manual
Change-Id: Ie9b8a0d59611498909b5c78eca65fe563a65004b
Kept the lowest to highest ordering priority, under the assumption
that less has to change in the UI/consumer if the highest priority
is popped, since it will just take it off the end without shifting
any elements. The ordering is documented.
Bug: 184891031
Test: atest DomainVerificationManagerApiTest
Change-Id: I6ea5908e356ee96aa440650f3d5025319c195266
gets sent to
This adds an API to allow sending a broadcast to an app that has permission FOO but not permission BAR.
Earlier, a lack of this option was causing a double send problem for broadcasts, which should now be resolved.
Test: manual
Bug: 183537857
Change-Id: I13287d63e039164694be4daa832bb23709990054
+revert the Incremental installation API to be SystemApi as the corresponding native API is not exposed in the NDK.
Bug: 184713911
Fixes: 184713911
Test: atest PackageManagerShellCommandTest PackageManagerShellCommandIncrementalTest IncrementalServiceTest PackageManagerServiceTest ChecksumsTest
Change-Id: I466eb6d2f98aed75c22b92877d5e8cb3ddf839bc
The change makes the distinction between VibrationEffect and
CombinedVibratio clearer. The later is a combination of the former with
the extra information about the vibrator ids, allowing effects to be
played in one or more vibrators in parallel or in sequence.
The methods create/start synced where also renamed to parallel (together
with respective builder classes), to indicate they perform one or more
effects in parallel on multiple vibrators. This is also a better match
to the sequential combinations.
This change also deprecates the Context.VIBRATOR_SERVICE in favour of
the new VIBRATOR_MANAGER_SERVICE. The default vibrator can be retrieved
from the manager system service.
Same deprecation applied to InputDevice.getVibrator method.
Fix: 184123900
Test: CombinedVibrationTest
Change-Id: I44d8b225098d35fbf7783254acaf6a78f9fb4505
Currently, the Translation uses feature to determine if we need to
start the translation system service. But our case is like the
content capture not the autofill, the autofill can have the feature
without any service defined and the user can install one later. But
translation isn't, we should need to check config instead of
feature.
The feature will not be used anymore, it's safe to delete it. If we
leave the removal to next OS, it is painful to delete it.
Bug: 183360041
Test: atest CtsTranslationTestCases
Change-Id: Ib6886a17965937abf064e526c36c67428af7398f
1. Close AppSearchSession after use.
When the device is shutdown with open AppSearchSession, if the data is
not properly persisted to disk, it would lead to a recovery upon reboot,
which could hurt boot performance.
We cannot rely on shutdown signal to close existing session because the
broadcast is not guranteed. Since session creation is light-weight, it
is more performant to simply close the session everytime after use.
2. Refrain from calling setup schema on boot.
To improve the performance of device start up, we want to avoid calling
AppSearchSession#setSchema unnecessarily since it could hurt performance.
In this CL we saved the version of the schema in the xml file which can
serve as an indicator of whether schema setup is required.
Bug: 151359749
Test: atest ShortcutManagerTest1 ShortcutManagerTest2
ShortcutManagerTest3 ShortcutManagerTest4 ShortcutManagerTest5
ShortcutManagerTest6 ShortcutManagerTest7 ShortcutManagerTest8
ShortcutManagerTest9 ShortcutManagerTest10 ShortcutManagerTest11
ShortcutManagerTest12
Test: atest CtsShortcutManagerTestCases
Change-Id: I2d6b66cef0c152ad99611dec3dee32a473bef4a9
Reduce search space in AppSearch with following changes:
1. Introduce inverted flag upon saving document.
2. Implement AppSearch queries for more general cases.
Bug: 151359749
Test: atest ShortcutManagerTest1 ShortcutManagerTest2
ShortcutManagerTest3 ShortcutManagerTest4 ShortcutManagerTest5
ShortcutManagerTest6 ShortcutManagerTest7 ShortcutManagerTest8
ShortcutManagerTest9 ShortcutManagerTest10 ShortcutManagerTest11
ShortcutManagerTest12
Test: atest CtsShortcutManagerTestCases
Change-Id: I94ec6a2b04c83ef3689c65d63741340b01c9ff50
* changes:
Fix domain verify restore and add signature check
Support serializing package signatures for domain verification state
Add domain verification CTS to TEST_MAPPING
Fix DomainVerificationService deadlock
Check installed and enabled state for package domain approval
Revoke domain user selection when approved through shell
