New revisions should be affected changes of the reference field

The last submitted patch, 2: 3338387-2.patch, failed testing. View results →
- codesniffer_fixes.patch Interdiff of automated coding standards fixes only.

#3 had issues with translations. This patch addresses that by only comparing revision IDs if there isn't a translation.

#2's test-only patch is still valid so I'm not uploading another one of those.

#5 doesn't actually work in my use case. It works in the test because there's no translations. The test should include that.

Here's one that checks for the same revision ID if the langcode is the same (so it isn't true just because there is some translation, and it IS true if the revision for any translation has changed).

The test in this one does use translation, so it catches the problem with #5.

The last submitted patch, 7: 3338387-7-test-only.patch, failed testing. View results →
- codesniffer_fixes.patch Interdiff of automated coding standards fixes only.

The last submitted patch, 7: 3338387-7.patch, failed testing. View results →
- codesniffer_fixes.patch Interdiff of automated coding standards fixes only.

I can't access the old DrupalCI test results anymore and the Gitlab CI testing is broken currently for 8.x-1.x anyway, so not really sure that #9 is still relevant.

I hit this as well and the patch solved the issue perfectly, thanks a lot @travis-bradbury. Nice catch with the translation issue, as well! 👍

It also comes with test coverage which is also great.

RTBC from my point of view.

Only previous major ci is broken, merge requests are required now.

Well the CI fails, but not on the added test method, so I guess back to RTBC.

HEAD is passing, fully now and so was current major tests before, if this fails then it's because of this change. Maybe some expectations need to be adjusted and are currently wrong, but it is failing either way :)

Looked into this, and, yes, there this does break things when working with translated content. EntityReferenceRevisionsCompositeTranslationTest::testCompositePendingRevisionTranslation() creates an English node, then creates a pending revision for that, then a German translation and then a pending revision for the German translation. The node_field_revision table then (after line 232) looks like this:

With this patch it looks like this at the same point in the test:

I didn't investigate any further, because I now want to go back to see exactly the use-case why I needed the patch in the first place in the light of this finding. But wanted to share this in case this helps anyone else push this along or provide further insight.