This PR was merged into the 4.4 branch.
Discussion
----------
[Cache] Run Redis Sentinel tests in GithubAction
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
Add support for Sentinel tests in GHA
Commits
-------
bd84394dc5 Run Redis Sentinel tests in GithubAction
This PR was merged into the 4.4 branch.
Discussion
----------
[LDAP] Add ldap tests to github CI
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#39028
| License | MIT
| Doc PR |
Adds LDAP test on github actions pipeline and corrects sum bugs in the LDAP Component test
It also adds a bug resolution from @Nek- made in #38875
Commits
-------
ea78f728b1 Use GithubAction to run ldap tests
af9562b12c Adds LDAP Adapter test in integration group
This PR was merged into the 4.4 branch.
Discussion
----------
[Config] Recheck glob brace support after GlobResource was serialized
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
cc @bastnic
This bug was reported on Symfony Slack: `$this->globBrace` is set to `null` after unserialization from the `.meta` file.
Instead of serializing this property, I decided to reinitialize the property after unserialization. I think that's a safer option (e.g. it works when the cache is build on a different server with different globBrace support than the one running the application).
Commits
-------
d9534779cf Reinitialize globBrace after unserialization
This PR was merged into the 4.4 branch.
Discussion
----------
[HttpFoundation] Fix for virtualhosts based on URL path
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#34866
| License | MIT
| Doc PR | no
This PR fixes base URL detection when:
- virtualhost is based on URL path
- AND local path does not match that URL virtual host path prefix
fix covered with tests
Commits
-------
75ff86811f [HttpFoundation] Fix for virtualhosts based on URL path
This PR was merged into the 4.4 branch.
Discussion
----------
[Cache] Add tests on CacheDataCollector
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| License | MIT
<!--
Replace this notice by a short README for your feature/bugfix. This will help people
understand your PR and can be used as a start for the documentation.
Additionally (see https://symfony.com/releases):
- Always add tests and ensure they pass.
- Never break backward compatibility (see https://symfony.com/bc).
- Bug fixes must be submitted against the lowest maintained branch where they apply
(lowest branches are regularly merged to upper ones so they get the fixes too.)
- Features and deprecations must be submitted against branch 5.x.
-->
The `calculateStatistics()` method of `Symfony\Component\Cache\DataCollector\CacheDataCollector` contain a lot of logic and manipulate multi-dimensional array that could be refactor with VO.
But before doing this, I would add test on this part.
Commits
-------
7b4310f045 Add tests on CacheDataCollector
This PR was merged into the 4.4 branch.
Discussion
----------
[DependencyInjection] Fix circular detection with multiple paths
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#39056
| License | MIT
| Doc PR | -
There are currently 2 kind of issues related to the Dependency Injection:
1. performance issue when project contains many loops (#37850)
Which has been fixed by #38882
2. Infinity loop in some case (#38970)
Which has been fixed by #38980 and #39021
The new issue #39056 has been introduced by #38882 (The performance issue refactor) because in order to optimize loop detection, I take a short cut and choose to not collect ALL the circular loop but only the one that matters
I was wrong. All loops matters.
This PR fix my previous refacto to collect ALL the paths, with a low CPU footprint
Commits
-------
1c3721e8ad Fix circular detection with multiple paths
This PR was merged into the 4.4 branch.
Discussion
----------
[Console] Fix ANSI when stdErr is not a tty
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#38981
| License | MIT
| Doc PR | -
Taking the @wouterj 's comment into account (https://github.com/symfony/symfony/issues/38981#issuecomment-721915428)
This PR prevents using the same Formatter for stdOut and stdErr when possible.
When user send a custom formatter (or call `setFormatter`) the previous logic is kept.
Otherwise, symfony is asked to create the Formatter, and thus is able to clone the formatter.
In a future PR targeting 5.3, we could improve the constructor to let people inject 2 distinguished formatters
Commits
-------
f3a398b5af Fix ANSI when stdErr is not a tty
This PR was merged into the 4.4 branch.
Discussion
----------
[DependencyInjection] Fix circular reference with Factory + Lazy Iterrator
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#38970
| License | MIT
| Doc PR | -
The issue, occurs when a `factory` iterates over services (think tagged iterator) that also need the `factory`.
The PhpDumper is not able to detect the loop because the TaggedService iterator is flaged as "lazy" which is ignored in the loop detection. 2d7e0b02c6/src/Symfony/Component/DependencyInjection/Dumper/PhpDumper.php (L474-L476)
See test case for a reproduce case.
This PR takes into account lazy services when computing loops.
I'm not sure this is the right thing to do /cc @nicolas-grekas .
A better solution could be to do this ONLY when the service is used as a factory?
Commits
-------
51ff060603 Fix circular referene with Factory and LazyIterator
This PR was merged into the 4.4 branch.
Discussion
----------
[PhpUnitBridge] fix replaying skipped tests
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
Spotted while working on https://github.com/symfony/polyfill/pull/310
See also https://3v4l.org/D6Gro for the newly skipped tests in symfony/intl
Commits
-------
849d1b3845 [PhpUnitBridge] fix replaying skipped tests
This PR was merged into the 4.4 branch.
Discussion
----------
[HttpKernel] Fix session initialized several times
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
The session injected in the request is a factory. When the request is **cloned** (ie sub-request) while the session is not yet used in the master request (the session is still a factory), both Master and Sub request will have a factory.
If both requests attempt to read the session afterward, they both triggers the factory, and both initialize the session.
It's not a dead end for the `SessionListener` because the session come from the container and is shared, but:
- the session is initialized ($storage->setOptions(['cookie_secure' => true])) twice
- if we replace the sesssion from container to a factory, it could be an issue
Commits
-------
30a3c7c87b Fix session called initized several time
This PR was merged into the 4.4 branch.
Discussion
----------
[DependencyInjection] Improve performances in CircualReference detection
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#37850
| License | MIT
| Doc PR | -
This PR change the way circular references are detected. And improve the project submitted in #37850 [by **86%**](https://blackfire.io/profiles/compare/cc4aa41d-5b63-4caa-9a1b-fc282dc5d64a/graph) (Build the container in 1.1 sec instead of 10)
Issue is: When a project contains a lot Circular Reference, the Dumper [spend a lot of time](https://blackfire.io/profiles/54f6eba2-b93c-4f2b-a268-1e58883faecb/graph) in merging those circular references.
note: a circular reference is not an issue when an service is not injected by constructor, but this Can not be known until all references are resolved (performed previously by connectCircularReferences)
This PR removed the connectCircularReferences and generate a flatten tree of dependencies:
- the key is the service ID
- the value is the list of direct **AND** indirect dependency + path to join the dependency
I also [benched the PR with a project with few references](https://blackfire.io/profiles/compare/2f9902e6-3347-40b3-8421-e1fd09c067d2/graph) and result are almost the same before/after.
Commits
-------
d4db75692b Improve performances in CircualReference detection
This PR was squashed before being merged into the 4.4 branch.
Discussion
----------
[Process] Dont test TTY if there is no TTY support
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no
| Deprecations? | no
| Tickets | Fix#38946
| License | MIT
| Doc PR |
Commits
-------
e918e5ab30 [Process] Dont test TTY if there is no TTY support
This PR was squashed before being merged into the 4.4 branch.
Discussion
----------
Fixing some Mongolian
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no
| Deprecations? | no
| Tickets | Fix#38916
| License | MIT
| Doc PR |
Hey @luffy1727, could you help me with the last two strings here?
Commits
-------
5c848d9796 Fixing some Mongolian
This PR was merged into the 4.4 branch.
Discussion
----------
[Validator] Merge RecursiveValidatorTest with its parents
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no
| Deprecations? | no
| Tickets | N/A
| License | MIT
| Doc PR | N/A
`RecursiveValidatorTest` as two abstract parent classes that to my understanding exist mainly for historic reasons: We used to have different validator implementations and those abstract classes contained some shared test cases. This is not the case anymore, which leaves `RecursiveValidatorTest` as the only child class of those two abstract classes.
This PR suggests to merge the three classes into one.
I've executed the test suite before and after the change and PHPUnit reported the same number of assertions and test cases.
I've also checked how the merge to 5.1 and 5.x would look like. The merge to 5.1 works without conflicts. From 5.1 to 5.x, two additional test cases have to be moved down and the reference to a few fixture classes needs to be adjusted.
Commits
-------
6822774d37 [Validator] Merge RecursiveValidatorTest with its parents.
This PR was merged into the 4.4 branch.
Discussion
----------
[Finder] Force set access time in test
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no, it only updates the tests
| New feature? | no
| Deprecations? | no
| Tickets | Fix#38942
| License | MIT
| Doc PR |
OSX don't really update atime (access time). They claim it is because performance reasons. So this PR will update the test to make PHP force access time on a file.
The different runs of `testAccept` depends on each other because each test modifies the filesystem a bit. That is why I added the extra `sleep(1)`.
Commits
-------
c5ec51abdf [Finder] Force set access time in test
