Failing test in module's existing coverage prevents unrelated patch submissions passing tests

I expect this patch to fail on 9.5 (the default), but pass on 10.1.

Do not merge this patch please :)

The last submitted patch, 2: pathauto-3390330-no-op-1.patch, failed testing. View results →

I tried to reproduce this failure against 9.5 locally, and was unable to.

From the test results in #2 above we can see that it appears to fail on Drupal 9.5, and pass on Drupal 10.1.

From https://www.drupal.org/pift-ci-job/2705047 → this test passed for Drupal 9.5 in June 2023.

Locally I see that the failing test passes in Drupal 9.5.

I'm not sure how to identify why this fails against 9.5 in DrupalCI currently. What am I missing?

I wouldn't bother with fixing that, if you can't reproduce then that's likely a DrupalCI thing. If you'd like to contribute, that time is likely better spent on switching to GitlabCI per https://www.drupal.org/docs/develop/git/using-gitlab-to-contribute-to-dr... → .

I'm fine with switching to D10.1 as the default branch, D9 EOL is 2 month or so.

Delighted to oblige! 📌 Switch to Gitlab CI Needs review

I've changed the default to 10.1, 9.5 will still run as a weekly test.

For the record, according to the weekly tests with postgresql ( https://www.drupal.org/pift-ci-job/2773503 → ), this started failing between the 14th and 21th September, either due to a Drupal core or Gitlab CI change.

FWIW I see also a couple fails at Drupal\Tests\pathauto\FunctionalJavascript\PathautoUiTest:102 back in May 2023 on the Postgres results → .

(I'm going to stop thinking about this now.)

I've hijacked the no-op patch in this issue for the purposes of running tests to confirm that a failure I was seeing in 🐛 Duplicate alias entities created with 'Create a new alias. Leave the existing alias functioning' setting RTBC when testing with MySQL 8 was not on account of the patch in that issue.

I've added these tests to the patch in #2 here, which I expected to all fail:

* PHP 8.2 & MySQL 8, D10.1
* PHP 8.1 & MySQL 8, D10.1
* PHP 7.4 & MySQL 8, D10.1

(I'm surprised I was permitted to use that last combo, so that will probably fail more comprehensively than the other two.)

Automatically closed - issue fixed for 2 weeks with no activity.