Combine fields filter REGEXP causes PostgreSQL syntax error

Nice find, we need a test for this, I've already written one in 🐛 A number of Views combined fields filter operators are currently pointless to an end user Needs work that just hasn't landed yet.

Probably easiest to grab the REGEX test from that and add it here and see if it indeed fails on PostGreSQL

I've attached two patches, one with the REGEX test and one with both REGEX test + fix.

The Needs Review Queue Bot → tested this issue.

While you are making the above changes, we recommend that you convert this patch to a merge request → . Merge requests are preferred over patches. Be sure to hide the old patch files as well. (Converting an issue to a merge request without other contributions to the issue will not receive credit.)

I've attached two patches, one with the REGEX test and one with both REGEX test + fix.

Nice, moved to a MR and lets see if the test run better there, works fine locally

PostgreSQL test-only fails as expected: https://git.drupalcode.org/issue/drupal-3415118/-/jobs/653536

So this looks good to me

Hiding patches

The testbot for PostgreSQL is still failing. See: https://git.drupalcode.org/issue/drupal-3415118/-/jobs/653536

@daffie the visual representation leaves a little to be desired to make it clear, but if you look at the top level Postgres build you can see that the normal test runs pass and only the test-only run fails, as we would hope

https://git.drupalcode.org/issue/drupal-3415118/-/pipelines/79102

if you look at the top level Postgres build you can see that the normal test runs pass and only the test-only run fails, as we would hope

I did not know we had a "tests only" run. That is really great!
@lendude: Thank you for the explanation!

@ahmad.k, Welcome to Drupal!

I'm triaging RTBC issues → . I read the IS and the comments. At first, this is a bit confusing because the standard template was not used and there is no statement of the problem. But all the information is there.

I read the MR and the test. It all appears it order. And it too a few clicks but I have confirmed that a test run on PostgreSQL.

Leaving at RTBC;

Committed and pushed b3c74ca14e to 11.x and c42d624654 to 10.3.x and f0e32bb3f3 to 10.2.x. Thanks!

Backported to 10.2.x as a non-disruptive bug fix.

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