This PR was merged into the 2.2 branch.
Discussion
----------
[Serializer] Fixed the error handling when decoding invalid XML to avoid a Warning
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a
| License | MIT
| Doc PR | n/a
Without the exception being thrown here, the call to ``simplexml_import_dom`` throws a warning saying ``Invalid Nodetype to import``.
This has been reported by @Tobion after we refactored FOSRestBundle to use this XmlEncoder instead of a duplication of a previous version of the class: https://github.com/FriendsOfSymfony/FOSRestBundle/pull/583/files#r7221045
Commits
-------
b2550b9 Fixed the error handling when decoding invalid XML to avoid a Warning
This PR was merged into the 2.2 branch.
Discussion
----------
[DomCrawler] [HttpFoundation] Make `Content-Type` attributes identification case-insensitive
According to [section 3.7 of RFC 2616][], media-type attribute names in the `Content-Type` header are case-insensitive.
Therefore, identification of the `text` type and the `charset` parameter in the `Content-Type` header should be case-insensitive.
[section 3.7 of RFC 2616]: http://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html#sec3.7
Commits
-------
17a2d66 [DomCrawler] [HttpFoundation] Make `Content-Type` attributes identification case-insensitive
This PR was merged into the 2.2 branch.
Discussion
----------
[2.2][Process] Fix#9343 : revert file handle usage on Windows platform
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #9343
| License | MIT
Hello,
I propose to revert the use of file handles only for `STDOUT` on Windows platform (see implementation in 2.2.6 [here](4059720232/src/Symfony/Component/Process/Process.php (L231-L242))).
When I decoupled pipes management from `Process` in #8924, I used file handles for both `STDOUT` and `STDERR`. This was an error as it introduced random failure in reading the handles (reported as [PHP#65650](https://bugs.php.net/bug.php?id=65650)).
Reverting to the previous implementation solves the issue. My apologies for the issues it introduced.
Versions that have been affected by the bug are 2.2.7, 2.2.8, 2.2.9, 2.3.4, 2.3.5 and 2.3.6.
Side note : I thought about testing the file handles implementation on *nix, but it fails most of the time where as Windows is okay. Unit testing on windows is okay (AbstractProcessTest::testProcessPipes tests it), but I don't provide a travis compatible test.
Commits
-------
e9dd408 [Process] Fix#9343 : revert file handle usage on Windows platform
This PR was submitted for the master branch but it was merged into the 2.2 branch instead (closes#9372).
Discussion
----------
Fixed English usage for existing users.
Commits
-------
b674e44 Fixed English usage for existing users.
This PR was merged into the 2.2 branch.
Discussion
----------
[2.2][Form] Improved FormTypeCsrfExtension to use the type class as default intention if the form name is empty
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
ping @stof
follow-up PR to #9327
Commits
-------
219e44d [Intl] Improved FormTypeCsrfExtension to use the type class as default intention if the form name is empty
This PR was merged into the 2.2 branch.
Discussion
----------
[DoctrineBridge] Added type check to prevent calling clear() on arrays
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Commits
-------
4ff59d7 [DoctrineBridge] Added type check to prevent calling clear() on arrays
This PR was submitted for the master branch but it was merged into the 2.2 branch instead (closes#9312).
Discussion
----------
[Security] Fix docblock typo
Commits
-------
bff65b2 [Security] Fix docblock typo
This PR was merged into the 2.2 branch.
Discussion
----------
[2.2][Form] Changed FormTypeCsrfExtension to use the form's name as default intention
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Before, every form used the same "intention"/"csrf_token_id" value by default, namely "unknown". This PR fixes the default value to the form's name, which is at least different for forms with (a) explicit names and (b) different types, where the implicit name equals the type's name.
Commits
-------
b07c618 [Form] Changed FormTypeCsrfExtension to use the form's name as default intention
This PR was merged into the 2.2 branch.
Discussion
----------
[DoctrineBridge] Loosened CollectionToArrayTransformer::transform() to accept arrays
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Previously, writing an association getter like this was impossible:
```php
public function addTag(Tag $tag) { ... }
public function removeTag(Tag $tag) { ... }
public function getTags()
{
return $this->tags->toArray();
}
```
Using `toArray()` is a useful way to restrict modifications of the collection to the specified methods. But previously, CollectionToArrayTransformer failed in this case, because it did not accept arrays as input.
Commits
-------
55001ab [DoctrineBridge] Loosened CollectionToArrayTransformer::transform() to accept arrays
This PR was merged into the 2.2 branch.
Discussion
----------
Dead code v2.2
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | ~
| License | MIT
| Doc PR | ~
Commits
-------
0946be6 Removed unused use statements.
This PR was submitted for the master branch but it was merged into the 2.2 branch instead (closes#9283).
Discussion
----------
[Security] Added japanese translation resource for security component.
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
Commits
-------
4176cf1 Added japanese translation resource for security component.
This PR was merged into the 2.2 branch.
Discussion
----------
[Yaml] Fixed the escaping of strings starting with a dash when dumping
Dashes need to be escaped in character sets in regexes as they are used to specify a range otherwise.
It fixes the dumper bug reported in #9039 (the parsing issue is not fix here)
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | partially #9039
| License | MIT
| Doc PR | n/a
Commits
-------
af369ae [Yaml] Fixed the escaping of strings starting with a dash when dumping
This PR was submitted for the master branch but it was merged into the 2.2 branch instead (closes#9270).
Discussion
----------
[Templating] Fix in ChainLoader.php
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
The LoaderInterface requires the _time_ parameter to be passed.
Commits
-------
2631eb3 [Templating] Fix in ChainLoader.php
This PR was submitted for the master branch but it was merged into the 2.2 branch instead (closes#9246).
Discussion
----------
[Session] fixed wrong started state
Currently the session stays in the started mode after the call of ```save()``` which will close the session. We found the issue in our custom SAPI which keeps the symfony stack initialized over multiple requests.
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Commits
-------
d641cd6 [Session] fixed wrong started states
This PR was submitted for the master branch but it was merged into the 2.2 branch instead (closes#9237).
Discussion
----------
[FrameworkBundle] assets:install command should mirror .dotfiles (.htaccess)
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
The `assets:install` command currently ignores all *.dotfiles* when mirroring the `Resources\public` folders of bundles. This can lead to issues when *.htaccess* files are required for some public assets (ie.: CORS headers for font files to be served via Cloudfront).
Since the assets being installed are clearly in a folder called `public`, we can safely assume that those files are in fact supposed to be accessible and copy them over with the normal files.
Commits
-------
9c884a0 [FrameworkBundle] assets:install command should mirror .dotfiles (.htaccess)
This PR was submitted for the master branch but it was merged into the 2.2 branch instead (closes#9223).
Discussion
----------
[Translator] PoFileDumper - PO headers
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #6765 - partially
| License | MIT
| Doc PR |
Po files should not be dumped without a header. It causes a lot of problems, including charset issues.
See #6765 / 1 for more info
Commits
-------
e93bc50 [Translator] PoFileDumper - PO headers
This PR was merged into the 2.2 branch.
Discussion
----------
[Process][2.2] Fix 9182 : random failure on pipes tests
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #9182
| License | MIT
I'm not a big fan of this fix, but - at least - it works.
With this code, finally, Process does not behave the same at all on Windows and Linux.
This patch does not smells very good but it solves the random failing test issue (that produced at runtime too).
Actually, calling `proc_get_status` within the waiting loop introduced the bug.
So this PR reverts to the previous behavior (consider a process running as long as pipes give data). On Windows, this is not the same behavior as we're not using streams but file handles. Whereas the feof of a stream is detected when the other side closes, the feof of a file handle can be reached at any time. So, on Windows, `proc_get_status` is called (checking the feof of the file handle might be positive until the executable outputs something), and we consider a process running as long as the information returned says it's running.
We could think of decouple windows and linux logic in two separated objects, using the interfaces I introduced in #8753. This could bring much more readability and make the code more easy to understand.
Commits
-------
64a0b40 [Process] Fix#9182 : random failure on pipes tests
This PR was merged into the 2.2 branch.
Discussion
----------
[Bridge] [Propel1] Fixed guessed relations
<table>
<tr><th>Q</th><th>A</th></tr>
<tr><td>Bug fix</td><td>yes</td></tr>
<tr><td>Feature addition</td><td>no</td></tr>
<tr><td>Backwards compatibility break</td><td>no</td></tr>
<tr><td>Symfony2 tests pass</td><td>yes</td></tr>
<tr><td>Fixes the following tickets</td><td>N/A</td></tr>
<tr><td>Todo</td><td>N/A</td></tr>
<tr><td>License of the code</td><td>MIT</td></tr>
<tr><td>Documentation PR</td><td>N/A</td></tr>
</table>
The `PropelTypeGuesser` did not match OneToMany relations properly. For example if you have a `Author` class with `Comments` attributes the guesser would guess "Comment" instead of "Comments".
I added some tests to ensure the non regression.
Commits
-------
0b1e95f Fixed propel guessed relations
This PR was merged into the 2.2 branch.
Discussion
----------
[FramworkBundle] Check event listener services are not abstract
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
Commits
-------
6744ead [FramworkBundle][HttpKernel] Check event listener services are not abstract
This PR was submitted for the master branch but it was merged into the 2.2 branch instead (closes#9207).
Discussion
----------
[HttpKernel] Check for lock existence before unlinking
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
My logs are filled with a bazillion errors stating "Warning: unlink(/var/www/mysite/app/cache/prod/http_cache/md/cf/47/c693da5dab3eccb65fa36a9b4b07ad0f7cc4.lck): No such file or directory in /var/www/mysite/vendor/symfony/symfony/src/Symfony/Component/HttpKernel/HttpCache/Store.php line 53"
Commits
-------
89cd3b9 Check for lock existence before unlinking
My logs are filled with a bazillion errors stating "Warning: unlink(/var/www/mysite/app/cache/prod/http_cache/md/cf/47/c693da5dab3eccb65fa36a9b4b07ad0f7cc4.lck): No such file or directory in /var/www/mysite/vendor/symfony/symfony/src/Symfony/Component/HttpKernel/HttpCache/Store.php line 53"
This PR was merged into the 2.2 branch.
Discussion
----------
Fixed cache warmup of paths which contain back-slashes
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #8781
| License | MIT
| Doc PR | n/a
From #8781
"
At this step in the warmup process there is a string replacement made, but paths with backslashes
are written into the cache file as something like `C:\\path\\to\\cache/pro_/...` but the str_replace
searches for the plain string `C:\path\to\cache/pro_/...`, which produces no matches.
The fix solves this by using var_export to escape the search (and replace) paths so they match the
format originally output into the cache file.
"
Commits
-------
526d285 [FrameworkBundle] fixed path replacement on Windows
