Twitter had a set of Compose rules, which have been forked into Detekt, Ktlint, and Android Lint implementations.
I see ktlint as mostly being formatting, Detekt being more for multiplatform static analysis, and Android Lint for Android-specific issues. Android lint is slow, so I didn’t want to use it as the first choice.
I went with the Detekt implementation which can be useful if we leverage multiplatform for our Composables in the future.
* [#653] Move copy to buffer action
- Trigger this action after seed phrase panel click and confirm via dialog window
- Added basic ui tests
- Added also dialog integration test
- Added related strings
* Revert "[#653] Move copy to buffer action"
This reverts commit 813eab00b747a779be5ef652745002f65c04572c.
* [#150] Refactoring the Backup flow to use Compose Scaffold
* Fix Backup flow screenshot test
- Removed scroll actions above nodes, which are no longer part of scroll behaviour - bottom navigation buttons are now part of Compose Scaffold component.
* Added scroll actions in screenshot test of Profile screen
- After tested the whole app with screenshot test on smaller screen device 4WVGA Nexus S
* Remove unnecessary screenshot test click action
- This click action on the Profile screen title seems to be unnecessary for the test and creates confusion
* ScreenshotTest module auto components init
- Changed the way we auto-initialize components in ScreenshotTest module
- Now we use androidx-startup library for it
- And we disabled the default way
* Add system back button navigation support
* Enable scrolling for Backup Test screen
* Fix Screenshot test on small screen in landscape
- Tested and fixed cases in which our screenshot test wasn't successful
- Tested on 4 WVGA Nexus S in landscpae mode
* Code clean
* Address review comments + stages refactoring
- Flattened BackupStage sealed class. Test and Failure are now regular stages.
- Introduced CheckSeed stage, instead of reusing Seed phrase for re-viewing.
- Simplified BackupView and custom Saver implementations.
- List of possible screen state stages is now lazy loaded value, instead of method.
- Some of the stages now override stage moving methods to enhance Backup screen state machine, as it's not linear.
- Existing tests updated to align with the new implementation of stages.
* Remove `run` block
* Rename CheckSeed -> ReviewSeed
Check might imply to another reader of the code that there is some going back to the test. I went with the word Review which seems to better convey how that screen is passive for the user.
* Simplify list construction
This should have better performance.
* Crash instead of allowing back navigation
* Add documentation
* Fix initialization error
* Add non-localized string tag
Co-authored-by: Carter Jernigan <git@carterjernigan.com>
* [#667] Fix for Update VM flaky test results
- Added small delays to the AppUpdateCheckerMock functions to copy a real-world situation a bit more
- This also helps with satisfying expectations in UpdateViewModelTest
* Typo fix
* Make duration more explicit
Co-authored-by: Carter Jernigan <git@carterjernigan.com>
* Temporary enable FTL test
* Enable emulator.wft tests for ui-integration-test module on CI
* Rebase onto main
* Test Fladle configuration for ui-integration-test module
* Fix Fladle for ui-integration-test module
* Rename ui-integration-test module flade configuration
* Disable again FTL action from PRs
* Clear support for FTL from ui-integration-test module
* Refactor the emulator.wtf support across our modules.
* [#599] Macrobenchmark test module
- Created new dedicated macrobenchmark test module
- Updated related Architecture and Setup documentation
- Connected to app module
- New benchmark build type
- Related run configuration above custom Gradle task
- Basic startup benchmark test included
* Benchmark build variant simplification
* Run benchmarking simplification
* Documentation update.
* New property IS_SIGN_RELEASE_BUILD_WITH_DEBUG_KEY
- It enables signing the release build with debug key configuration. Default value is false.
- First, we check if we can sign it with release configs, otherwise with debug.
- Documentation updated.
* Benchmarking documentation update
* Adds support for Android SDK 32 and 33
- Bumped benchmark library to the latest alpha version
- Which results in a need of adding profile-installer library in target module (app)
- Now we're able to run benchmarking on Android SDK level 29 and higher (i.e. latest levels included)
- Updated related documentation
* Enables benchmarking for emulators
- I've decided to support emulator devices with our benchmark test module at the end. Documentation and code comments explain to a user that the results may not be the same as on a real physical device. But until we have it set up in CI, we can benefit from having it set up like this.
* [#525] Refactoring Integration Test Module
- Change module type to test + related Gradle scripts change
- Rename module to - remove the trailing Lib suffix + documentation
- Move module structure from test to main source
* Static checks result fix
* Temporary enable FTL test
* Test updated emulator.wft version
* Update locked emulator.wtf dependency version
* Enable emulator.wft tests for ui-integration-test module on CI
* Fix enable emulator.wft tests for ui-integration-test module on CI
- Update documentation
* Test Fladle configuration for ui-integration-test module
* Fix Fladle for ui-integration-test module
* Rename ui-integration-test module flade configuration
* Disable again FTL action from PRs
* Clear support for FTL from ui-integration-test module
* Refactor the emulator.wtf support across our modules.
* Fix Mermaid graph syntax
* Minor change in comment
* Fix ui-integration-test module run configuration
- Fixed manually
- Aligned with other run configurations
- Stable and beta versions of AS behave differentially, but after this fix it seems to be working as expected in both
* Update emulator.wtf run configuration
- Covered ui-integration-test module too
- Changed configuration name
Carter Jernigan
Alex
Honza Rychnovsky