Come together with the global Drupal community in Rotterdam, 28 Sept – 1 Oct 2026. Sessions, contribution, connection, and Early Bird savings until 8 June.Problem/Motivation
Drupal\Tests\locale\Functional\LocaleUpdateTest::testUpdateImportSourceRemote
Behat\Mink\Exception\ResponseTextException: The text "Updates for: Contributed module one, Contributed module two, Custom module one, Locale test" was not found anywhere in the text of the current page.
/builds/project/drupal/vendor/behat/mink/src/WebAssert.php:811
/builds/project/drupal/vendor/behat/mink/src/WebAssert.php:262
/builds/project/drupal/core/modules/locale/tests/src/Functional/LocaleUpdateTest.php:136
/builds/project/drupal/vendor/phpunit/phpunit/src/Framework/TestResult.php:728
https://git.drupalcode.org/project/drupal/-/jobs/164513
Steps to reproduce
Proposed resolution
Remaining tasks
User interface changes
API changes
Data model changes
Release notes snippet
| Comment | File | Size | Author |
|---|---|---|---|
| #6 | 5000x_testUpdateImportSourceRemote.patch | 19.88 KB | spokje |
| #6 | 5000x_testUpdateImportSourceRemote-fix.patch | 20.48 KB | spokje |
| #2 | Available translation updates Drupal.htm_.txt | 22.81 KB | spokje |
Comments
Comment #2
spokjeAttached the HTML output of the failed test in the IS.
Test looks for "Updates for: Contributed module one, Contributed module two, Custom module one, Locale test"
Page shows: " Updates for: Locale test, Custom module one, Contributed module one, Contributed module two"
Comment #3
spokjeComment #4
spokjeThis fails "random-consistently" only on pgsql.
Since that's tested "only" daily, not on commit, it seems like this one flew under the radar for a while.
If we switch to GitLab CI for realz, we might want to do an on-commit check on all main supported DBs?
Comment #5
catchWe have that set up, but I think we possibly broke it, re-opened #3392739: _TARGET_DB_TYPE does not exist for now.
Comment #6
spokjeThanks @catch.
The problem in this fail seems to be that we don't sort the list with locale updates on name, only on language.
So it seems we were "saved" by almost all DB-envs almost always returning them in the order we liked, instead in the order we assured it to be.
Comment #7
spokjeHmmm, seems like this is not only a cosmetic problem in sorting on the page, but a more serious problem in ordering or not importing .po-files...
Comment #8
spokje