#Needs tests

Just figured out that the failing of:

• Drupal\Tests\package_manager\Build\PackageInstallTest
• Drupal\Tests\package_manager\Build\PackageInstallSubmoduleTest
• Drupal\Tests\package_manager\Build\PackageUpdateTest

with the error message
Some modules have database schema updates to install. You should run the database update script immediately
was caused by the incorrect placement of the post_update hook function in the link.module file instead of the required link.post_update.php file.

Attached a static inline_entity_form--2025-01-04--mr-126.patch file, from the MR126 up to this point.
to be used with Composer Patches

Could someone clarify if using the MR!10504 is enough to replace the Layout Builder Symmetric Translations → functionality?

I'm using Drupal 10.4.0 right now and feel exhausted fighting with layout_builder_st problems, currently, it's not allowing me to clear caches from the Drupal admin interface.

Thanks.

With test coverage removed seems pretty straight forward fix then.

May need to think approach as not seeing why we are adding comment hooks into node. Should try and keep things separate if possible.

The tests seem fixed now, pipeline runs all green except for the test-only test: fails like it should.

That seems to leave several out-of-scope typehints that need to be removed from the MR.

Threads have been resolved.

Read the documentation and reads fine to me and will agree with the CR not being needed so closed that thread too.

Think this one is ready.

Left comments.

I don't remember why we thought contextual links did not make sense. I think it is because the use cases are relatively complex. Not having contextual links for the simple use cases in core was ok for the moment. When embedding a document link, the contextual links seem like a bit of overkill, but when embedding a slideshow, or something complex like that, it totally makes sense.

My first feeling is it depends on the view mode / media type whether contextual links make sense. At least per media type. Can we change the configuration of the filter to allow which media types/view modes should have contextual links?

Maybe we should even move the configuration to the media type / view mode config so we can also use this for entity references for example?

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

There's also filter.format.basic_html.yml files in standard and demo_umami profiles that will need to be updated.

Small comment but also needs rebase

Unnecessary test removed and rebased MR.

the test can be removed here. it's testing that something we removed was removed

• Updated Issue Summary.
• Rebased the MR over the most recent 11.x branch.
• Resolved all GitLab threads.
• Added a post_update hook and a test for it.

The MR Gitlab pipeline is failing and I believe it has to do with some unrelated problem in the 11.x branch. Not setting this issue to "Needs Review" until this is solved.

vidorado → changed the visibility of the branch 11.x to hidden.

Rebased and addressed feedback

I like the simplistic implementation from @kingdutch, so I have used that as base and made it configurable.
The MR 164 covers the following:
- Invalidate tokens by specific actions (multiple can be selected) via settings.
- Available actions can be altered by swapping out/decorating ```TokenExpiryTriggerHandler` service``` + ```Oauth2TokenSettingsForm```.
- Default actions are set so we don't introduce BC.

Todo: add test coverage.

We had a similar issue in DS, see 🐛 Config schema issues with extended DS layouts Active to allow layouts having additional configuration. Any tips always welcome, although I don't think there is one :)

saving credits

added more more comment that I missed before.

@smustgrave, I have addressed the comments on the MR, please check.

Ok, newer contrib modules, newer core, now on 10.4.0 with PHP 8.3, this issue came back !

Had to roll a new patch!

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

Thanks @douggreen. Confirming to work well on 10.4.0

I applied this patch locally and it worked for me. The change made to google_analytics.js in 4.0.3 was pretty small, just adding a conditional: https://git.drupalcode.org/project/google_analytics/-/commit/b698b97cc55...

Please check my work.

I can confirm that this patch fixes the issue on Drupal 10.4.0 as well.

Re-roll.

Posting the latest diff as a patch.

Comment on MR.

Maybe we can skip deprecation test since defineOptions isn't used anywhere. But the current test I don't believe is actually testing anything, just throwing a deprecation but no assertion.

Probably needs usability sign off

We have failing tests, and still need new tests. The MR also introduces PHPCS issues.

Re-rolled for 11.x

Test status - Pass
Verified an issue on Drupal 11.X version for Olivero and claro themes. MR applied successfully. Issue looks fixed.

Followed steps below-
1. login as admin
2. use claro or olivero
3. goto => structure => block layout
4. switch to mobile device you will see contents are overflowing outside page from right.

Attaching screenshots for reference.

RTBC+1

The Views handler doesn’t require a sort option, which is the root cause of the issue.

I’m hoping this doesn’t require any tests? It’s just a quick DX fix to help people get over their migrations. I'm RTBC'ing because it gets me (and others) over the hump.

Here's my bad data:
a:8:{s:12:"entity_types";a:0:{}s:17:"field_permissions";a:1:{s:4:"type";i:0;}s:7:"indexes";a:1:{s:9:"target_id";a:1:{i:0;s:9:"target_id";}}s:8:"settings";a:4:{s:7:"handler";s:5:"views";s:16:"handler_settings";a:2:{s:9:"behaviors";a:1:{s:17:"views-select-list";a:1:{s:6:"status";i:0;}}s:4:"view";a:3:{s:4:"args";a:2:{i:0;s:56:"[field_collection_item:host:node:field-research-area:id]";i:1;s:47:"[field_collection_item:host:node:field-year:id]";}s:12:"display_name";s:17:"entityreference_1";s:9:"view_name";s:24:"utility_er_courses_upper";}}s:16:"profile2_private";b:0;s:11:"target_type";s:8:"academic";}s:12:"translatable";i:0;s:7:"storage";a:5:{s:4:"type";s:17:"field_sql_storage";s:8:"settings";a:0:{}s:6:"module";s:17:"field_sql_storage";s:6:"active";s:1:"1";s:7:"details";a:1:{s:3:"sql";a:2:{s:18:"FIELD_LOAD_CURRENT";a:1:{s:37:"field_data_field_draft_course_section";a:1:{s:9:"target_id";s:36:"field_draft_course_section_target_id";}}s:19:"FIELD_LOAD_REVISION";a:1:{s:41:"field_revision_field_draft_course_section";a:1:{s:9:"target_id";s:36:"field_draft_course_section_target_id";}}}}}s:12:"foreign keys";a:1:{s:12:"eck_academic";a:2:{s:5:"table";s:12:"eck_academic";s:7:"columns";a:1:{s:9:"target_id";s:2:"id";}}}s:2:"id";s:2:"70";}

And here's how I tracked it down:
SELECT * FROM `field_config` WHERE (`module` = 'entityreference') AND (`data` NOT LIKE '%sort%')

@anybody, that makes sense since we were actually scanning for broken links so it's possible those pages were missing assets. Possible connection with search_api however in our case the site in question had very little done using search_api , possibly didn't have anything done with search_api.

A kernel test needs to be created to verify the fix.

Believe the kernel test failure is relevant here.

I haven’t written tests before, but I want to learn how to do so. I’ll give this task a try and keep you updated on my progress.

Left some comments on MR

Thanks

prem suthar → made their first commit to this issue’s fork.

Thanks you @shalini_jha for building the test.

Thank you @smustgrave, @quietone and @catch for moving this along.

Ok I address your comments and rebased, I fixed one more typo as well.

I'll keep an eye on tests and move status after.

Great thanks! I applied most suggestions, I'll rebase and address the rest of your comments later today or early tomorrow.

@berdir is right this is one of two blockers to deprecate hook hook info which is the real target.

Fixed the two remarks, English is not my native language 😇

Don't really think we need a CR as it was also internal before?

> @nicxvan, why did you tag this as 11.2.0 release priority?

Because this is a blocker to deprecate hook_hook_info() usage, as it relies on those extra include files being loaded.