[RFC] DDC-3976 #1551
[RFC] DDC-3976 #1551
Conversation
|
Hello, thank you for creating this pull request. I have automatically opened an issue http://www.doctrine-project.org/jira/browse/DDC-3990 We use Jira to track the state of pull requests and the versions they got |
guilhermeblanco
force-pushed
the
DDC-3976
branch
from
163e594
to
13d1ec2
Compare
guilhermeblanco
force-pushed
the
DDC-3976
branch
from
13d1ec2
to
84151ec
Compare
guilhermeblanco
force-pushed
the
DDC-3976
branch
from
af6f94a
to
373c9cd
Compare
guilhermeblanco
force-pushed
the
DDC-3976
branch
from
373c9cd
to
367e90f
Compare
| @@ -96,6 +96,26 @@ public function testPutAndLoadManyToOneRelation() | |||
|
|
|||
| $this->assertEquals($this->states[1]->getCountry()->getId(), $c4->getCountry()->getId()); | |||
| $this->assertEquals($this->states[1]->getCountry()->getName(), $c4->getCountry()->getName()); | |||
|
|
|||
| //evict collection on add | |||
| $prev = $c3->getCities(); | |||
There was a problem hiding this comment.
Please split into a separate test
|
Except for the test note, 👍 |
| $this->collectionUpdates[$coid] = $value; | ||
| } | ||
|
|
||
| $this->collectionUpdates[$coid] = $value; |
There was a problem hiding this comment.
Isn't this making also the inverse side of collections checked? :-\
There was a problem hiding this comment.
Yes, but you're indeed updating a collection adding/removing new items, so it should be checked depending of CollectionPersister implementation.
In most cases, it'll just return, except on CacheCollectionPersister, which would evict the collection (addressing the bug)
| @@ -67,6 +67,7 @@ public function buildCacheEntry(ClassMetadata $metadata, CollectionCacheKey $key | |||
| foreach ($collection as $index => $entity) { | |||
| $data[$index] = new EntityCacheKey($metadata->name, $this->uow->getEntityIdentifier($entity)); | |||
| } | |||
|
|
|||
There was a problem hiding this comment.
Extra newline in unrelated file?
guilhermeblanco
force-pushed
the
DDC-3976
branch
2 times, most recently
from
e60f7aa
to
58992ad
Compare
|
Hi @Ocramius @guilhermeblanco , i was just looking at the code, and i found out that this commit never reached the branch 2.5. Is there any reason for this? |
|
Possibly didn't manage to backport, or @guilhermeblanco decided that the fix didn't need backporting or is too much effort to do so. |
|
@Ocramius is there any guide to know how you guys are doing the backport? I would like to help to do it. |
|
@nicolas-cajelli don't have a guide, but here's what I do: git checkout 2.5
git checkout -b a-new-branch-that-backports-fix-#1234-to-2.5
git cherry-pick aaaaaaa1
git cherry-pick aaaaaaa2
git cherry-pick aaaaaaa3
git push origin a-new-branch-that-backports-fix-#1234-to-2.5Then open a PR from that branch to 2.5 |
This release relaxes [`doctrine/common`](https://github.com/doctrine/common) requirements in order to allow installation of versions that support PHP 7.1 features in proxy class generation. Please note that a similar requirement relaxation still needs to be applied to [`doctrine/dbal`](https://github.com/doctrine/dbal) in order to allow installation of the latest `doctrine/common` versions. [doctrine#6156](doctrine#6156) This version also backports some fixes around the eviction of the second level cache entries of inverse side associations in one-to-many - many-to-one mappings. [doctrine#6159](doctrine#6159) Further fixes were applied in order to have child classes in inheritance mapping share the same timestamp region when the second level cache is enabled. [doctrine#6028](doctrine#6028) Also, `Doctrine\ORM\EntityManager#merge()` now triggers `Doctrine\ORM\Events::prePersist` listeners with the merged entity state whenever an internal `Doctrine\ORM\UnitOfWork#persist()` call is implied. [doctrine#6177](doctrine#6177). Total issues resolved: **8** - [5570: Fix PrePersist EventListener when using merge instead of persist](doctrine#5570) - [6028: Make child entity share the timestamp region with parent class](doctrine#6028) - [6110: Clear $this->collection even when empty, to reset keys](doctrine#6110) - [6156: Allow doctrine/common 2.7](doctrine#6156) - [6159: doctrine#5821 Backport doctrine#1551 - Fixed support for inverse side second level cache](doctrine#6159) - [6174: Merging a new entity with PrePersist event make changes in callback not be considered](doctrine#6174) - [6177: Fix doctrine#6174 doctrine#5570: merging new entities should also trigger prepersist lifecycle callbacks with merged entity data](doctrine#6177) - [6178: Backport doctrine#6177 - fix doctrine#6174 doctrine#5570: merging new entities should also trigger prepersist lifecycle callbacks with the merged data](doctrine#6178) # gpg: directory `/c/Users/PC/.gnupg' created # gpg: new configuration file `/c/Users/PC/.gnupg/gpg.conf' created # gpg: WARNING: options in `/c/Users/PC/.gnupg/gpg.conf' are not yet active during this run # gpg: keyring `/c/Users/PC/.gnupg/pubring.gpg' created # gpg: Signature made Tue Dec 20 00:49:05 2016 FLEST using DSA key ID 12EC2DF8 # gpg: Can't check signature: public key not found # Conflicts: # lib/Doctrine/ORM/Event/LoadClassMetadataEventArgs.php # lib/Doctrine/ORM/Tools/Pagination/LimitSubqueryOutputWalker.php # tests/Doctrine/Tests/ORM/Functional/QueryCacheTest.php # tests/Doctrine/Tests/ORM/Tools/Pagination/LimitSubqueryOutputWalkerTest.php # tests/Doctrine/Tests/ORM/Tools/Pagination/PaginationTestCase.php
Fixes #1542