The 2026 Drupal Business Survey is open. Your response will help shape the future of the Drupal business ecosystem.Problem/Motivation
Follow-up to #2565887: Fix test failures stemming from Dynamic Page Cache
Proposed resolution
Remaining tasks
User interface changes
API changes
Data model changes
| Comment | File | Size | Author |
|---|---|---|---|
| #7 | 2566537-cache-7.patch | 2.69 KB | tim.plunkett |
| #2 | 2566537-route-cache-2.patch | 8.78 KB | tim.plunkett |
| #2 | interdiff.txt | 4.02 KB | tim.plunkett |
| #2 | 2566537-route-cache-2-refactoring-only.patch | 6.13 KB | tim.plunkett |
Comments
Comment #2
tim.plunkettSplit this into two hunks. First is the refactoring, second is the fix for disabling. Not yet working for deleting.
Also needs web tests.
Comment #3
tim.plunkettComment #4
dsnopekShouldn't we be able to clear the
'page_manager_page:' . $path->id()cache tag rather than having to look up the route name? Because it's the same cache entry's that'll get removed (assuming I understand how that works correctly). Or is it a render cache entry that isn't tagged with the page manager page? (And if so, where could that be coming from?)Comment #5
berdirYes, if we make sure that the response has the page_manager cache tag, then delete should just work. Which is already the case for block display variant, we'll have to do it for the response object too. Which might need a custom one, since we have that additional response interface now.
Or actually, that should work too, through the 4xx-response cache tag, that is invalidated on every entity save?
Comment #6
tim.plunkettOh duh, that's why I couldn't get this to break in tests, it was just using the HTTP status variant!
Comment #7
tim.plunkettHm, now I can't reproduce this at all. Is there something wrong with my test?
Comment #8
tim.plunkettNot sure what happened here, need to revisit or close out.
Comment #9
japerryPage Manager 8.x-1.x is no longer supported, closing.