This PR was merged into the 2.3 branch.
Discussion
----------
[Form][Validator] All index items after children are to be considered grand-children when resolving ViolationPath
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | unsure, see note below
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #11458
| License | MIT
| Doc PR | -
#### Possible BC Break
The old behavior had unit test cases specifically testing the case of a grand-children form. However, this behavior is not documented anywhere and the fix seems to have no adverse effects on form validation. `Symfony\Component\Form\FormInterface` implements `ArrayAccess`, therefore, semantically speaking, `children[direct_child].children[grand_children]` and `children[direct_child][grand_children]` are equivalent. `offsetGet` is expected to fetch an element from `children`. I do not see why both were not considered equivalent when resolving the ViolationPath.
This commit will indeed change how some errors are mapped. However since the old mapping is (in my opinion) a bug...
Commits
-------
c64a75f [Form][Validator] All index items after children are to be considered grand-children when resolving ViolationPath (fixes#11458)
This PR was merged into the 2.3 branch.
Discussion
----------
[Form] FormBuilder::getIterator() now deals with resolved children
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
I think FormBuilder::getIterator() should resolve children before makes an iterator because it seems to be used in same purpose with FormBuilder::all().
What do you think?
Commits
-------
0deb505 [Form] FormBuilder::getIterator() now deals with resolved children
This PR was squashed before being merged into the 2.6-dev branch (closes#11917).
Discussion
----------
[Validator] Add ClassMetadata plural methods for convinience
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #4143
| License | MIT
| Doc PR | -
I realised there's no specific place to document this methods, as the code examples always include all the formats. I think it's enough if IDE autocompletes these methods.
Commits
-------
0fd6769 [Validator] Add ClassMetadata plural methods for convinience
This PR was merged into the 2.3 branch.
Discussion
----------
[SwiftmailerBridge] Bump allowed versions of swiftmailer
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes (but SwiftmailerBridge itself does not contain any tests)
| Fixed tickets | -
| License | MIT
| Doc PR | -
Current version of Swiftmailer is 5.2.1, while (previously to this commit)
the version installed by composer was 5.0.3.
This is rather important, since 5.2.1 closes a security issue that 5.0.3 is
vulnarable to (https://github.com/swiftmailer/swiftmailer/issues/494).
Commits
-------
a5e3fc9 [SwiftmailerBridge] Bump allowed versions of swiftmailer
This PR was merged into the 2.3 branch.
Discussion
----------
[DependencyInjection] remove `service` parameter type from XSD
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR | symfony/symfony-docs#4222
Referencing a service in a parameter doesn't work and will lead to an error when the configuration is loaded (see symfony/symfony-docs#4211).
Commits
-------
7333c2d remove `service` parameter type from XSD
This PR was merged into the 2.3 branch.
Discussion
----------
[Intl] Updated icu.ini up to ICU 53
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Extracted from #9206.
Commits
-------
260e2fe [Intl] Updated icu.ini up to ICU 53
This PR was merged into the 2.3 branch.
Discussion
----------
[Intl] Added exception handler to command line scripts
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Extracted from #9206.
Commits
-------
9052efc [Intl] Added exception handler to command line scripts
This PR was merged into the 2.3 branch.
Discussion
----------
[Intl] Removed non-working $fallback argument from ArrayAccessibleResourceBundle
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
The code in question didn't actually work. This was extracted from #9206.
Commits
-------
5feda5e [Intl] Removed non-working $fallback argument from ArrayAccessibleResourceBundle
This PR was submitted for the master branch but it was merged into the 2.3 branch instead (closes#11497).
Discussion
----------
Use separated function to resolve command and related arguments
Hi,
This PR split command and related arguments resolution into two distinct functions.
It will help to solve the HHVM issue sensiolabs/SensioDistributionBundle#150 .
Thanks,
Jérémy
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | yes
| BC breaks? | yes
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | ~
| License | MIT
| Doc PR | ~
Commits
-------
ee75af0 Use separated function to resolve command and related arguments
Current version of Swiftmailer is 5.2.1, while (previously to this commit)
the version installed by composer was 5.0.3.
This is rather important, since 5.2.1 closes a security issue that 5.0.3 is
vulnarable to (https://github.com/swiftmailer/swiftmailer/issues/494).
This PR was merged into the 2.3 branch.
Discussion
----------
[Security] Use hash_equals for constant-time string comparison (again)
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a
| License | MIT
| Doc PR | n/a
Use the `hash_equals` function (introduced in PHP 5.6) for timing attack safe string comparison when available.
Add in the DocBlock that length will leak (https://github.com/symfony/symfony/pull/11797#issuecomment-53990712).
Commits
-------
3071557 [Security] Add more tests for StringUtils::equals
03bd74b [Security] Use hash_equals for constant-time string comparison
This PR was merged into the 2.6-dev branch.
Discussion
----------
[Bundle][FrameworkBundle] make the stopwatch service always available
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #11347
| License | MIT
| Doc PR |
Previously, one had to be careful to check if the ``debug.stopwatch`` service was available before using it. Otherwise, the application would break in the prod environment.
Commits
-------
ffc4090 make the stopwatch service always available
This PR was merged into the 2.3 branch.
Discussion
----------
[DI] Added safeguards against invalid config in the YamlFileLoader
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #11333
| License | MIT
| Doc PR | n/a
Exceptions explaining the mistake are better than fatal errors or weird notices appearing when trying to deal with such invalid data.
The XML file loader is not affected by this because the data are validated with the XSD before being processed
Commits
-------
5183501 [DI] Added safeguards against invalid config in the YamlFileLoader
This PR was submitted for the 2.5 branch but it was merged into the 2.3 branch instead (closes#11897).
Discussion
----------
[FrameworkBundle] Remove invalid markup
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #11896
| License | MIT
| Doc PR | n/a
Commits
-------
1fe8e31 [FrameworkBundle] Remove invalid markup
This PR was merged into the 2.3 branch.
Discussion
----------
[Intl] Added "internal" tag to all classes under Symfony\Component\Intl\ResourceBundle
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | yes?
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
We didn't have this tag yet when this component was first written. The code in that
namespace is only used for resource bundle generation and was never meant for public
use.
We need to include in the update notes that users should check for usage of these classes.
Commits
-------
7fd5e8b [Intl] Added "internal" tag to all classes under Symfony\Component\Intl\ResourceBundle
We didn't have this tag yet when this component was first written. The code in that
namespace is only used for resource bundle generation and was never meant for public
use.
This PR was merged into the 2.3 branch.
Discussion
----------
[FrameworkBundle] improve handling router script paths
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
The `server:run` command switches the working directory before starting the built-in web server. Therefore, the path to a custom router script had to be specified based on the document root path and not based on the user's working directory.
Another option is to update the documentation (as started in symfony/symfony-docs#4194). Though I think the current behaviour is a bug. The intended behaviour can be derived from the command's help message:
> ```
If you have custom docroot directory layout, you can specify your own
router script using --router option:
> ./app/console server:run --router=app/config/router.php
```
As you can see, the path is specified based on the current working directory.
Commits
-------
0a16cf2 improve handling router script paths
This PR was merged into the 2.6-dev branch.
Discussion
----------
[Monolog] added NotFoundActivationStrategy from MonologBundle
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a
| License | MIT
| Doc PR | n/a
The NotFoundActivationStrategy class from MonologBundle is totally independent of the framework and should be part of the bridge instead. That would allow people to use it easily with Silex for instance.
ping @Seldaek
Commits
-------
b064d2f [Monolog] changed the not found activation strategy to use the request stack
1a239af [Monolog] added NotFoundActivationStrategy from MonologBundle
This PR was submitted for the master branch but it was merged into the 2.3 branch instead (closes#11868).
Discussion
----------
Remove routes for removed WebProfiler actions
The import/export functionality was moved to commands in f38536ab79, but the routes were not removed.
Commits
-------
1421449 Remove routes for removed WebProfiler actions
This PR was merged into the 2.6-dev branch.
Discussion
----------
[FrameworkBundle][Translation] moved cache to Translation component (new PR)
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #10628
| License | MIT
| Doc PR |
This supersedes the PR #11197 (while including the changes made by it). I removed the `$options` argument for `Symfony\Component\Translation\Translator` and replaced it by a public method `enableCache($cacheDir, $debug = false)`. It aims to solve what @fabpot said about passing an array of options in #11197 <s>while not modifying the existing constructors</s>.
Commits
-------
30fed6a [Translation][Cache] Removed the options from the arguments of Translator
8b2d9a8 [FrameworkBundle][Translation] moved cache to Translation component
This PR was submitted for the master branch but it was merged into the 2.3 branch instead (closes#11860).
Discussion
----------
[Security] Fix usage of unexistent method in DoctrineAclCache.
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #10328
| License | MIT
| Doc PR |
The method `deleteByPrefix` does not exist. I replaced it by `deleteAll`: as @guilhermeblanco said, this method is not available in the interface `Cache` but it is present in the abstract class `CacheProvider`.
Commits
-------
131abd8 [Security] Fix usage of unexistent method in DoctrineAclCache.
Fixed phpdoc
Aligned variables and description
Removed enableCache and added cache setup in constructor
Added tests for locales with . and @ with caching
This PR was merged into the 2.3 branch.
Discussion
----------
[FrameworkBundle] backport more error information from 2.6 to 2.3
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #11724
| License | MIT
| Doc PR |
Commits
-------
87449e0 backport more error information from 2.6 to 2.3
The commit on master was:
server:run command: provide more error information
The server:run command didn't provide many information when the executed
command exited unexpectedly. Now, the process' exit code is passed through
and an error message is displayed.