Commits
-------
37bbd0f Moved symfony/config from the "recommend" dependency to the "suggest" dependency. Cannot find "recommend" in composer documentation
Discussion
----------
Moved symfony/config from the "recommend" dependency to the "suggest" dependancy
Moved symfony/config from the "recommend" dependency to the "suggest" dependency. Cannot find "recommend" in composer documentation
---------------------------------------------------------------------------
by igorw at 2012-07-21T10:52:19Z
Recommend used to exist but was removed.
👍
Commits
-------
3075fa6 [OptionsResolver] Renamed filters to normalizers
Discussion
----------
[OptionsResolver] Renamed filters to normalizers
Bug fix: no
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
Fixes the following tickets: -
Todo: -
This PR fixes the naming to be in line with the Serializer component.
Add Response as possible return type of the method because the method AbstractAuthenticationListener::handle() test if $returnValue is an instance of Response (line 148).
Commits
-------
4eb54a0 update CHANGELOG
db9ea09 [Doctrine] [Bridge] fix repositoryMethod test
2a6c222 Add a customRepository option to the uniqueEntity validator
Discussion
----------
[Doctrine] [Bridge] Add a "repositoryMethod" option to the uniqueEntity validator
Bug fix: no
Feature addition: yes
Backwards compatibility break: no
Symfony2 tests pass: yes
Fixes the following tickets: ~
Todo: ~
License of the code: MIT
Documentation PR: ~
This allows to configure the repository method used to verify uniqueness of entity.
Before, it was always using `findBy`.
---------------------------------------------------------------------------
by fabpot at 2012-07-20T05:35:28Z
Can you add a note in the CHANGELOG?
---------------------------------------------------------------------------
by docteurklein at 2012-07-20T07:17:08Z
@fabpot done.
Commits
-------
ed8823c [HttpFoundation] Allow setting an unknown status code without specifying a text
Discussion
----------
[HttpFoundation] Allow setting an unknown status code without specifying...
... a text
fix#4978
Commits
-------
16a980b [Validator] Fix bug order for constraint, property, getter and group-sequence-provider in validation.xml
Discussion
----------
[Validator] Fix bug order for constraint, property, getter and group-seq...
Actually, there is a bug that force developers to write validation.xml file with the following nodes order:
- constraint
- property
- getter
So that's not possible to have the following XML (because I need to write my property(ies) first).
```xml
<?xml version="1.0" encoding="UTF-8" ?>
<constraint-mapping xmlns="http://symfony.com/schema/dic/constraint-mapping"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://symfony.com/schema/dic/constraint-mappinghttp://symfony.com/schema/dic/constraint-mapping/constraint-mapping-1.0.xsd">
<class name="Application\Eko\MyBundle\Entity\MyEntity">
<getter property="isBar">
<constraint name="True">
<option name="message">My error message</option>
</constraint>
</getter>
<property name="foo">
<constraint name="NotBlank" />
</property>
</class>
</constraint-mapping>
```
The XML below result in the following exception:
```
[ERROR 1871] Element '{http://symfony.com/schema/dic/constraint-mapping}property': This element is not expected. Expected is ( {http://symfony.com/schema/dic/constraint-mapping}getter ). (in /var/www/myproject/src/Application/Eko/MyBundle/Resources/config/validation.xml - line 14, column 0)
```
This is due to the sequence element that needs to respect the order given in the schema file.
The choice element is doing the same thing and permit to have a free order of elements so I have replaced the sequence by a choice element.
For more information: http://www.w3.org/TR/xmlschema-0/#ref17
Commits
-------
c81b2ad [Form] Rename UnmodifiableFormConfig to ImmutableFormConfig
274eb9e [EventDispatcher] Rename UnmodifiableEventDispatcher to ImmutableEventDispatcher
Discussion
----------
Rename unmodifiable to immutable
Maybe it's just me, but it sounded really wrong. The EventDispatcher one was added in 2.1 so no BC break. I don't know about the Form one, but I guess it's just used internally anyway.
Commits
-------
39157a8 [Security] fixes multiple overlapping definitions of DefaultFailureHandler and DefaultSuccessHandler in AbstractFactory
Discussion
----------
[Security] fixes multiple overlapping definitions of DefaultFailureHandler and DefaultSuccessHandler in AbstractFactory
If more than one listener extends AbstractFactory, you'll have multiple calls to createAuthenticationFailureHandler and createAuthenticationSuccessHandler with the same id.
Implicitly it's going to use the one generated by the last factory generating unexpected behavior.
This is related to commits 915704c071 and c6aa392df7
Commits
-------
f5e03a3 Changed $request->setDefaultLocale() to $request->setLocale() in custom LocaleListener.
Discussion
----------
Changed $request->setDefaultLocale() to $request->setLocale() in custom ...
...LocaleListener.
---------------------------------------------------------------------------
by mpiecko at 2012-07-19T13:05:07Z
This is based on my comments from https://github.com/symfony/symfony/pull/4692#issuecomment-7096255
Commits
-------
310c458 [Process] Fixed a problem on RHEL5 where the exit code was incorrect
Discussion
----------
[Process] Fixed a problem on RHEL5 where the exit code was incorrect
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
License of the code: MIT
RHEL5 will intermittently result in an exit code of -1 due to `proc_get_status()` being called after the process has completed but outside of `updateStatus()` which saves the exit code.
See composer/composer#876
Commits
-------
3bb975c [WebProfilerBundle] Fixed white pixel in toolbar info
Discussion
----------
[WebProfilerBundle] Fixed white pixel in toolbar info
Fixed this 
Tested on IE9/FF/Chrome
RHEL5 will intermittently result in an exit code of -1 [1] due to
proc_get_status() being called after the process has completed
but outside of updateStatus() which saves the exit code.
[1]: https://github.com/composer/composer/issues/876
Commits
-------
96638f4 [UPGRADE] Tweaking formatting error with event listener section. Also tweaking values
Discussion
----------
[UPGRADE] Tweaking formatting error with event listener section and information
Hey guys!
This is just a documentation change based on sha: 5da1bc6a4f
Check out the details in the commit message - I'm not sure if this is right, but the original commit seemed a bit odd compared to what I was seeing in the code. There was also a fix to the formatting.
If I've missed something, I'm happy to change the commit as needed.
Thanks!
---------------------------------------------------------------------------
by stof at 2012-07-14T21:53:12Z
There is no equivalent to early_kernel_request for the router listener as both methods have been merged together (it was separated so that the initialization of the RequestContext was done before the firewall, but the whole logic is before the firewall now)
Commits
-------
89975ef Added more verbose message for exception when form types have wrong getName method
Discussion
----------
Added more verbose message for exception when form types have wrong getName
PR based on discussion of commit introduces exception when custom form type have no getName method. 6489a65960
---------------------------------------------------------------------------
by biozshock at 2012-07-17T08:15:35Z
@tyx thanks. Fixed. ;)
---------------------------------------------------------------------------
by fabpot at 2012-07-17T20:46:57Z
Can you squash your commits before I merge? Thanks.
---------------------------------------------------------------------------
by noetix at 2012-07-17T23:49:51Z
I think the variables should have quotes around them. It seems like all other error messages have them.
From: The type name specified for the service form.type.translatable does not match the actual name. Expected translatable_field, given translator
To: The type name specified for the service "form.type.translatable" does not match the actual name. Expected "translatable_field", given "translator"
---------------------------------------------------------------------------
by biozshock at 2012-07-18T09:39:01Z
@fabpot Done.
@noetix Thanks for remark.
Commits
-------
1f33756 Update master
Discussion
----------
Allow HttpKernel ->render to be called directly
Hi,
I faced this problem in a funny situation.
Working on tests of individual components, I want to be able to test sub-requests. These elements are not mapped directly, relying on `FrameworkBundle` routing internal to achieve its goal.
The nature of my application leads to fully decoupled bundles, being a website responsible to define everything, from app configuration to routing elements. That way, an individual bundle controller don't know the route it should call, leading the test to be harder to do.
Together with this situation, it is also required in my testing scenario to be able to test the real world execution, then being an ESI include (a sub-request). This test then needs to be able to directly call controller to be rendered. That said, `HttpKernel` provides an API that does exactly what is required to achieve my goal, calling `render()` which triggers the sub-request and also accepts options which allows me to test the real world scenarios.
But as soon as you trigger the method, `ProfileListener` intercepts the kernel response to collect profiling information. Under this specific situation, since you called directly `render`, there's no master request, then leading the test to fail with the following PHP warning:
```
Warning: SplObjectStorage::offsetExists() expects parameter 1 to be object, boolean given in /var/www/nde/vendor/symfony/symfony/src/Symfony/Component/HttpKernel/EventListener/ProfilerListener.php line 120
```
To fix that, all that is needed is a check for a possible parent request. That's the purpose of this patch. =)
---------------------------------------------------------------------------
by stof at 2012-07-17T15:11:13Z
This looks good to me, but as said on IRC, a test should be added to avoid regressions
