Problem/Motivation
Resizing a window in \Drupal\Tests\layout_builder\FunctionalJavascript\BlockFormMessagesTest::testValidationMessage() appears to fix a random fail.
Looking at the html again - we have a Placeholder for the "New title" block but we're expecting the block to have been replaced real block and be identified by #layout-builder .block-system-powered-by-block
Proposed resolution
For whatever reason, the screen resize isn't necessary anymore and the test doesn't have random failures without it.
Let's remove the resize so the test is inline with (almost) every other FunctionalJavascript test.
Remaining tasks
User interface changes
None
API changes
None
Data model changes
None
Release notes snippet
N/a
| Comment | File | Size | Author |
|---|---|---|---|
| #14 | 3055982-14.patch | 847 bytes | spokje |
Comments
Comment #2
alexpottPostponed on #2901792: Disable all animations in Javascript testing
Comment #3
krzysztof domański#2901792: Disable all animations in Javascript testing has been committed.
Comment #12
spokjeI believe the screen resizing isn't necessary anymore.
The normal routine to prove a random failure is fixed is to run the failing patch and the patch with the fix at the same time, whilst the latter has to have ~5000-8000 failure free runs to prove it's credibility.
Since this issue is a bit different, there is no failing test, we can "just" run the patch without the screen resize and see if it has ~8000 - 10.000 failure free runs
So let's do that here.
Comment #13
spokjeOk...
5000x seems a bit too much, after we hit around 45 minutes in the test run, things go "wonky".
Let's see what 3000x brings us
Comment #14
spokje11 * 3000 = 33.000 runs without a failure.
Good enough for me, here's the actual patch.
Comment #15
spokjeComment #16
spokjeComment #17
borisson_This issue is to figure out why this line was making the test pass, but something in the meanwhile has made this no longer needed. The patch in #14 looks great. I would love to know why but I think we can commit this as-is.
Comment #19
longwaveCommitted c61764b and pushed to 11.x. Thanks!
Wary of backporting things that might cause random fails, it's not hurting anything so let's put this in 11.x only.
Comment #20
longwave