## `integration_ui_keyboard_resize` flakiness reduction
This PR addresses flakiness in the `integration_ui_keyboard_resize` test
(issue #161300) and slightly improves its performance.
**Issue summary:**
The `integration_ui_keyboard_resize` test occasionally fails because the
device keyboard does not reliably open at the start of the test. This PR
mitigates one cause of this flakiness, though other potential underlying
issues may still exist. See #161300 for a detailed investigation.
**Changes and rationale:**
1. **Improved keyboard detection logic (flakiness reduction):**
* The core change moves `await driver.tap(defaultTextField);` *inside*
the loop that checks for layout changes. This ensures the test
repeatedly attempts to open the keyboard *while* waiting for the
expected layout shift, significantly improving reliability.
* **Local testing results (MacBook Pro M1 + Pixel 8 Pro):**
* **Original implementation:** Consistent failures within the first 15
iterations.
* **This PR:** Ran consistently for at least 300 iterations (out of
2000) before any failure, with some runs exceeding 900 iterations.
2. **`enableTextEntryEmulation` configuration improvement:**
* The `keyboard_resize.dart` app is *exclusively* used for testing and
requires `enableTextEntryEmulation: false` for proper device keyboard
interaction.
* This PR sets `enableTextEntryEmulation: false` directly within the
app's `enableFlutterDriverExtension` setup. This simplifies the test
setup, avoids cross-thread communication, and prevents the app from
appearing broken when run independently (as manual text input wouldn't
work otherwise).
3. **UI enhancements for manual testing:**
* Added `SafeArea` and `InputDecoration` to `keyboard_resize.dart`. This
ensures the `TextField` is fully visible and usable during manual
testing (it was previously obscured by the Android status bar). This
improves the developer experience when debugging the app directly.
4. **Optimized polling for speed:**
* Reduced the polling interval and increased the number of polling
iterations (maintaining the overall timeout). This change resulted in a
~2-second speed improvement per test iteration during local testing.
**Further flakiness improvement discussion and ideas**
See the discussion at https://github.com/flutter/flutter/issues/161300
**Testing and reproduction:**
To reproduce the original flakiness and verify the fix (reproducibility
may vary based on setup; see
https://github.com/flutter/flutter/issues/161300#issuecomment-2618952501):
1. **Disable automatic reboots:** Comment out the `await
checkForRebootRequired();` line in
9e273d5e6e/dev/devicelab/lib/framework/framework.dart (L240-L243)
2. **Disable automatic retries:** Set `static const int retryNumber =
0;` in
9e273d5e6e/dev/devicelab/lib/framework/cocoon.dart (L56-L57)
3. **Install `hyperfine`:** This is a command-line benchmarking tool
(e.g., `brew install hyperfine` on macOS).
4. **Connect a device/emulator:** Connect a physical Android device or
start an emulator.
5. **Run repeated tests:** Use `hyperfine` to run the test multiple
times and capture the output:
```bash
hyperfine -r 100 'dart bin/test_runner.dart test -t
integration_ui_keyboard_resize > log.txt'
```
This auto-formats all *.dart files in the repository outside of the
`engine` subdirectory and enforces that these files stay formatted with
a presubmit check.
**Reviewers:** Please carefully review all the commits except for the
one titled "formatted". The "formatted" commit was auto-generated by
running `dev/tools/format.sh -a -f`. The other commits were hand-crafted
to prepare the repo for the formatting change. I recommend reviewing the
commits one-by-one via the "Commits" tab and avoiding Github's "Files
changed" tab as it will likely slow down your browser because of the
size of this PR.
---------
Co-authored-by: Kate Lovett <katelovett@google.com>
Co-authored-by: LongCatIsLooong <31859944+LongCatIsLooong@users.noreply.github.com>
This adds a LICENSE file to the packages/flutter directory so that it can be found when building, and a test to make sure it exists in produced binaries.
* Update project.pbxproj files to say Flutter rather than Chromium
Also, the templates now have an empty organization so that we don't cause people to give their apps a Flutter copyright.
* Update the copyright notice checker to require a standard notice on all files
* Update copyrights on Dart files. (This was a mechanical commit.)
* Fix weird license headers on Dart files that deviate from our conventions; relicense Shrine.
Some were already marked "The Flutter Authors", not clear why. Their
dates have been normalized. Some were missing the blank line after the
license. Some were randomly different in trivial ways for no apparent
reason (e.g. missing the trailing period).
* Clean up the copyrights in non-Dart files. (Manual edits.)
Also, make sure templates don't have copyrights.
* Fix some more ORGANIZATIONNAMEs
Before this, we had several places where an isReleaseMode was defined, all with the same definition. This just makes it more broadly visible to allow our users to use it, as well as creating debug and profile versions, and adding a device lab test for it.
Since this is a const value, this makes it possible for a developer to easily mark blocks that can be removed at AOT compile time.
* Make CupertinoApp and MaterialApp both use WidgetsApp for Navigator
* Make CupertinoApp and MaterialApp const constructors
* Make WidgetsApp routes aware
* Update tests
This test fails consistently on mac2 and mac3 with the attached Moto G4
devices but passes consistently on other machines.
Adding a delay of 1s right after driver.connect() in setUpAll() causes
it to pass on the machines in question, which suggests a race condition.
Specifically it looks like connect returns the moment Flutter Driver
identifies that the isolate is up and running, but empirically it looks
like we start running the first test before the UI is actually up. This
triggers a failure wherein we start looking for elements before they're
onstage.
Link to viewport.dart:213 at HEAD:
b2b4665926/packages/flutter/lib/src/widgets/viewport.dart (L213)
Stack trace:
FlutterDriver waitFor should find text "present"
```
DriverError: Error in Flutter application: Uncaught extension error while executing waitFor: NoSuchMethodError: The getter 'visible' was called on null.
Receiver: null
Tried calling: visible
#0 Object.noSuchMethod (dart:core/runtime/libobject_patch.dart:46:5)
#1 _ViewportElement.debugVisitOnstageChildren. (package:flutter/src/widgets/viewport.dart:213:36)
#2 WhereIterator.moveNext (dart:_internal/iterable.dart:439:11)
#3 Iterable.forEach (dart:core/iterable.dart)
#4 _ViewportElement.debugVisitOnstageChildren (package:flutter/src/widgets/viewport.dart:214:8)
#5 _DepthFirstChildIterator._reverseChildrenOf (package:flutter_test/src/all_elements.dart:54:15)
#6 _DepthFirstChildIterator.moveNext (package:flutter_test/src/all_elements.dart:45:19)
#7 CachingIterable._fillNext (package:flutter/src/foundation/basic_types.dart:252:27)
#8 _LazyListIterator.moveNext (package:flutter/src/foundation/basic_types.dart:279:21)
#9 WhereIterator.moveNext (dart:_internal/iterable.dart:438:22)
#10 CachingIterable._fillNext (package:flutter/src/foundation/basic_types.dart:252:27)
#11 _LazyListIterator.moveNext (package:flutter/src/foundation/basic_types.dart:279:21)
#12 Iterable.isEmpty (dart:core/iterable.dart:449:33)
#13 Iterable.isNotEmpty (dart:core/iterable.dart:456:27)
#14 FlutterDriverExtension._waitForElement. (package:flutter_driver/src/extension/extension.dart:215:51)
#15 FlutterDriverExtension._waitUntilFrame (package:flutter_driver/src/extension/extension.dart:197:19)
#16 FlutterDriverExtension._waitForElement (package:flutter_driver/src/extension/extension.dart:215:11)
#17 FlutterDriverExtension._waitFor (package:flutter_driver/src/extension/extension.dart:286:11)
#18 FlutterDriverExtension.call (package:flutter_driver/src/extension/extension.dart:168:51)
#19 BindingBase.registerServiceExtension. (package:flutter/src/foundation/binding.dart:370:32)
```
Removes a previous hack that no longer appears to help (adding a 1
second delay in setUpAll() does seem to work around this issue though).
