[Validator] Fixed grouped composite constraints #29499
Conversation
HeahDude
force-pushed
the
validation-groups-collection
branch
from
3adf789
to
b53d911
Compare
|
Thanks @HeahDude. |
This PR was merged into the 3.4 branch. Discussion ---------- [Validator] Fixed grouped composite constraints | Q | A | ------------- | --- | Branch? | 3.4 | Bug fix? | yes | New feature? | no | BC breaks? | no | Deprecations? | no | Tests pass? | yes | Fixed tickets | #17675, #25888, #23480 | License | MIT | Doc PR | ~ From Lisbon :). Thanks @stof, @xabbuh for your help to finally fix this old issue. Commits ------- b53d911 [Validator] Fixed grouped composite constraints
|
I think this change completely obsoletes the caching. I am not sure, but a constraint that has two grps would now be validated twice, if the both grps are the validation target. |
|
Most constraints don't extend the |
|
Yes you are right, i was desperately searching for the check for |
|
Thanks for the fix @stof ! Just FYI, I think (I'm still investigating) this might have been a breaking change for us. Suddenly, some of our forms return unexpected validation errors. I think our code depended on this bug. Hopefully the fix should be quick. Nevertheless it is a bit surprising to see this in a patch release. I think we might have to check every single validation rule that uses collections in our project, because some previously unknown validations might come into action silently, preventing customers from doing things that were previously possible. |
From Lisbon :). Thanks @stof, @xabbuh for your help to finally fix this old issue.