I hope that this will lock bugs like b/298114205 in the test that created the bad condition, not the next one.
Bug: 298114205
Test: presubmit
Flag: N/A
Change-Id: If9b48fb5159fc35948607a99f87199d5a32e64d6
Renaming the following test names:
- testDragAppIconToWorkspaceCell to testDragAppIconToMultipleWorkspaceCells
- testDragShortcutToMultipleWoCells to testDragShortcutToMultipleWorkspaceCells
Bug: 298234173
Test: atest OopTaplDragTest
Change-Id: I16707e0a882164c7a5acec6a4197bd0bea8514a7
I hope that this will lock bugs like b/298114205 in the test that created the bad condition, not the next one.
Bug: 298114205
Test: presubmit
Flag: N/A
Change-Id: I6ad3598fb675d8b24ec0f273d255b7f2464dc53f
zhidou@ from ACE team is working on AConfigFakelib which will be useful
to set fake flag values in both unit tests and UI automator tests.
Meanwhile, we will mock the flag values for unit tests and UI automator
tests will have all flags values defaulted to false.
testAConfigLib with "test:true" doesn't generate fake flags values and
forces us to set the flag values using "SetFlagsRule". "SetFlagsRule"
doesn't work with Out Of Proc tests and hence doesn't serve the purpose
for us. AConfigFakelib and FakeFlagTargetPreparer should use for Out Of
Proc tests.
testAConfigLib with "test:true" is suitable for java host tests where
there is no device involved. Launcher tests are instrumentation tests and tests run on actual
devices. So we can use the production version of the AConfig lib and device
config flag values will be fetched for UI Automator tests.
Also migrating "enable_twoline_allapps" to trunk stable. Unit tests is
using mockito to mock the flag value as false. The owner of
"enable_twoline_allapps" can mock the value to true in unit tests as
next step.
Bug: 294913042
Test: Local studio build and soong build
Flag: enable_twoline_allapps
Change-Id: Iea42bf18197832a9ecbddc1c3aaa8c9e36169131
- Add mockito-inline-extended to allow mocking static methods.
- Add StaticMockitoRule to help mock static methods.
This helps in unit tests with mocking the trunk stable flags.
Bug: 294913042
Test: NA
Flag: NA
Change-Id: If53ae1d1bc68a5c6b1824b6c7060cbc14e095b1b
Use `enable_expanding_pause_work_button` as an example.
See go/trunk-stable-launcher for implementation details
Bug: 294913042
Test: adb shell device_config put launcher com.google.android.platform.launcher.aconfig.flags.enable_expanding_pause_work_button true
Flag: enable_expanding_pause_work_button
Change-Id: I732722c8b219c023adf5bf31f132ce9da72fc4d5
Now when exiting the test, we finish Launcher activity and force the leak check.
I've seen this helping one of the leaks to stay within the test that caused it instead of failing the next test.
Bug: 187761685
Flag: N/A
Test: presubmit, local
Change-Id: I7eac26f749433c37a6880a2f035ab1ba9c5fef6e
For now, only Alpha Jumps detector is implemented. The test will fail if
an alpha jump is detected.
The ignore-list for alpha jump detection ended up being uncomfortably
long, I plan dealing with this immediately after landing this CL.
Bug: 286251603
Test: presubmit, forrest
Flag: N/A
Change-Id: I3f44b02015e2ce93eb8b4071becbc44b36416459
For now, the rule is basically empty, to start with. In the following CLs it will
invoke the search for anomalies in the data collected by ViewCaptureRule
after the test succeeds and fail the test if an anomaly is found.
Bug: 286251603
Test: presubmit
Flag: N/A
Change-Id: I373ea0d7ee85206101fb15584889ea5933e7d7b8
This enables developers to watch failed tests in classes like
TaplTestsQuickstep in the go/web-hv tool.
Bug: 242867462
Test: Failed a test and verified that the TimeLapse bugreport entry
showed up properly in the go/web-hv tool.
Change-Id: Ic89af2a0e7102db52c52ddc668607a81c4809ed6
It also adds an assets folder inside of Launcher3 tests.
It also fixes a bug in fullReorder.
Test: atest ReorderWidgets
Flag: none
Bug: 229292911
Change-Id: I7dadd42a3057d353326691163bb7db71a2e9dcc4
This allows support for easily setting up default layouts
Bug: 277345535
Test: Presubmit
Flag: N/A
Change-Id: I1c089d60ac3f8add8d7e1060d343e04d30afe094
This is necessary because otherwise those are included in several gradle projects. Gradle doesn't support the same files imported in different projects, and it removes them from a random one if it happens, resulting in missing symbols.
This extracts the shared files in a new target, and adds it as dep of the original target.
Test: Build launcher + execute a few tests manually with gradle + verify that Launcher.java can resolve all symbols with gradle config
Bug: 262267728
Change-Id: Ida5b25c45320d517603834112f4699e4d3344c03
As now FlickerTests is leveraging TAPL to interact devices CUJs,
we realized in LauncherInstrumentation will verify task bar visibility
automatically to expect it always visible when quick-switch, which is
not always reliable since the task bar may be hidden by manual or
when the activity requests to show IME.
Add setIgnoreTaskbarVisibility in LauncherInstrumentation for the caller
to ignore taskbar visibility if the test does not need to verify it.
Bug: 228012334
Bug: 240306344
Test: atest FlickerTests:SwitchImeWindowsFromGestureNavTest
in tablet device
Test: atest NexusLauncherOutOfProcTests:com.android.quickstep.\
TaplTestsQuickstep#testQuickSwitchToPreviousAppForTablet
Change-Id: Id0a35561523d733b8434acb702ec7dcaa466a1c2
Navigation buttons take space on the hotseat, so hotseat width (qsb width and border space between icons) should be calculated instead of having a coded value.
Bug: 223724516
Bug: 228998463
Test: manual, changing display size
Test: manual, changing hotseatBarEndOffset value
Test: HotseatWidthCalculationTest
Change-Id: Ibd4f5ff2e06afda8e7420fb744db049d2e418e14
A test that takes > 3 min will generate an artifact file containing stacks of all threads of the test process taken every 3 sec. This artifact will be also generated if the test process is killed, for example, by timeout.
This artifact should help EngProd's effort to speed up presubmits.
Bug: 225186335
Test: local runs
Change-Id: I721779bfbe5bc6289315998ed2660f5f46165611
I want to include some stuff (like android.platform.test.rule.CrashDetector) but I want to land just the dependency part and see if I get postsubmit build problems that sometimes appear in obscure AOSP targets.
Test: presubmit
Bug: 187761685
Change-Id: I334b9f1029edbe8621df940900f43d0c5059373e
Merged-In: I334b9f1029edbe8621df940900f43d0c5059373e
When running on tablets in landscape, the QSB should inline with the hotseat icons. This does that and fixes small bugs that didn't account for the new hotseatBorderSpace.
Bug: 210118169
Test: atest Launcher3Tests:DeviceProfileTest
Test: visual, using HSV and Window
Change-Id: I5a89c5449bf59fde736618151eceaacca443ef67
1. Implement function to uninstall app from appIcon.
2. Add test uninstall to TaplTestsLauncher3.
Bug: 210927656
Test: Launcher3Tests:com.android.launcher3.ui.TaplTestsLauncher3#testTryUninstallFromWorkspace
Change-Id: Iedd3bd3a46bc626fc414fd8c5bd07ebc0fa235bb
In a former CL (http://ag/16064580) I've tried to
do the same but presubmit was constantly failing,
probably because the test is in kotlin.
Test: AddWorkspaceItemsTaskTest.kt
Bug: 199160559
Change-Id: Ie1bc4fcd4f94cd7cb0601c21bbdf273452b9dd1f
This will improve diagnostics for OOP tests,
like we now have a list of leaked activity classes.
Also some cleanups.
Bug: 187761685
Test: local runs
Change-Id: I8b5711ac727874fd826cfef9c742ea97048763e0
> Moving definition to bp files
> Adding mokito-inline to allow mocking final classes
> Moving src_common to src to avoid unnecessary code split
Bug: 196825541
Test: Presubmit
Change-Id: Ib6879cda6aee74f5b9050a16eb6ca94ed3b7e28c
Added SPDX-license-identifier-Apache-2.0 to:
quickstep/Android.bp
robolectric_tests/Android.bp
tests/Android.bp
Bug: 68860345
Bug: 151177513
Bug: 151953481
Test: m all
Exempt-From-Owner-Approval: janitorial work
Change-Id: I699e0796410f691ae7cf4802c41987d79e4e486d
