This PR was submitted for the 2.3 branch but it was merged into the master branch instead (closes#9578).
Discussion
----------
[DomCrawler] Fixes `attr` method returning empty string for missing attributes
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Bug discovered on Behat/MinkBrowserKitDriver#32.
Commits
-------
51d23fa [DomCrawler] Fixes `attr` method returning empty string for missing attributes
This PR was merged into the master branch.
Discussion
----------
[Debug] Fixed ClassNotFoundFatalErrorHandler which could cause a fatal error
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #9104
| License | MIT
| Doc PR | -
It might happen that `require_once` will include a file second time (for example with links or case-changed paths). Therefore we need to check if a class exists before requiring the file.
I also removed an unused argument from two methods (`$loader`).
Commits
-------
0baae4c [Debug] Fixed ClassNotFoundFatalErrorHandler which could cause a fatal error.
This PR was merged into the master branch.
Discussion
----------
Cache Warmup Breaks Namespaced Kernel
| Q | A
| ------------- | ---
| Bug fix? | [yes]
| New feature? | [no]
| BC breaks? | [no]
| Deprecations? | [no]
| Tests pass? | have not tried yet
| License | MIT
| Fixed tickets | further fixes#1431
My kernel has been moved and namespaced to `Cfs\Bundle\MultiSiteBundle\Kernel\CfsKernel`. This worked fine until a change was made to how the kernel temp stuff is handled in the warmup phase.
When the app generates its own cache (i.e you run cache without warmup and access the site) everything is generated ok and the .meta files generate the proper reference to the FQN of the Kernel.
However if the warmup is used, it uses `Cfs\Bundle\MultiSiteBundle\Kernel\CfsKerne_` as the temporary Kernel, and when it does "fix references to the Kernel in .meta files" it generates 2 errors.
1. It does not use a string safe tempKernel name, so it never finds the reference to the kernel
2. If you fix that, then it replaces the FQN of the tempKernel with `CfsKernel`, the non-namespaced name of the proper Kernel (it also leaves the character count wrong in the serialization `C:43:<class>` where 43 is the char count for the FQN above)
The two changes above fix this, by escaping the string and replacing it with a FQN Kernel Class name.
What are your thoughts on this? If this sounds reasonable i'll do further enhancements and check tests.
Commits
-------
9e7788e Cache Warmup Breaks Namespaced Kernel
It might happen that require_once will include a file second time (for
example with links or case-changed paths). Therefore we need to check
if a class exists before requiring the file.
This PR was merged into the master branch.
Discussion
----------
unify constructor initialization style throughout symfony
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | n/a
In almost all classes symfony uses property initialization when the value is static. Constructor initialization is only used for things that actually have logic, like passed parameters or dynamic values. IMHO it makes the code much more readable because property definition, phpdoc and default value is in one place. Also one can easily see what the constructor implements for logic like overridden default value of a parent class. Otherwise the real deal is just hidden behind 10 property initializations. One more advantage is that it requires less code. As you can see, the code was almost cut in half (210 additions and 395 deletions).
I unified it accordingly across symfony. Sometimes it was [not even consistent within one class](https://github.com/symfony/symfony/blob/master/src/Symfony/Component/Config/Definition/BaseNode.php#L32). At the same time I recognized some errors like missing parent constructor call, or undefined properties or private properties that are not even used.
I then realized that a few Kernel tests were not passing because they were deeply implementation specific like modifying booted flag with a custom `KernelForTest->setIsBooted();`. I improved and refactored the kernel tests in the __second commit__.
__Third commit__ unifies short ternary operator, e.g. `$foo ?: new Foo()`. __Forth commit__ unifies missing parentheses, e.g. `new Foo()`.
Commits
-------
077a089 unify missing parentheses
2888594 unify short ternary operator
2a9daff [HttpKernel] better written kernel tests
111ac18 unify constructor initialization style throughout symfony
This PR was merged into the master branch.
Discussion
----------
[HttpFoundation] fixed the issue described in #9480
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #9480
| License | MIT
| Doc PR | -
Commits
-------
7d35ce3 [HttpFoundation] fixed the issue described in #9480
This PR was merged into the master branch.
Discussion
----------
unify and fix callable phpdoc throughout symfony
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | n/a
I unified the phpdoc to use "callable" as type. It wasn't done consistently. It varied between mixed, callback, callable.
I think callable is best because php 5.4 even has a typehint callable.
Commits
-------
6d5ddce unify and fix callable phpdoc throughout symfony
This PR was merged into the master branch.
Discussion
----------
[FrameworkBundle] use the new request_stack service in the GlobalVariables object
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | ~
| License | MIT
| Doc PR | ~
Commits
-------
9c2ce49 [FrameworkBundle] use the new request_stack object in the GlobalVariables object
This PR was merged into the master branch.
Discussion
----------
[Console] Revert BC-break for verbose option value
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #9285
| License | MIT
| Doc PR | ~
Closes#9285
----
PR #8835 "fixed" the input value for verbosity option with `VALUE_OPTIONAL`. And syntax like
```
cli.php -v command
cli.php --verbose command
```
Is broken now, because `ArgvInput` tries to use argument as verbose value (#9285).
For now i added a test for this case and revert this changes.
Commits
-------
432bbe1 [Console] Revert a28eb8 because of BC break
5e03e9a [Console] Add test for --verbose before argument
This PR was merged into the master branch.
Discussion
----------
[Security] optimized ExpressionVoter
Commits
-------
a8c74d1 [Security] optimized ExpressionVoter
This PR was merged into the master branch.
Discussion
----------
[FrameworkBundle] use the new request_stack service to get the Request object in the base Controller class
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | ~
| License | MIT
| Doc PR | ~
Commits
-------
2e07338 [FrameworkBundle] use the new request_stack service to get the Request object in the base Controller class
This PR was merged into the master branch.
Discussion
----------
[Translation] make IdentityTranslater consistent with normal translator
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | n/a
The MessageSelector should not be required in the constructor of the IdentityTranslator because it's also optional for the [Translator](https://github.com/symfony/symfony/blob/master/src/Symfony/Component/Translation/Translator.php#L64).
I also fixed the TranslatorInterface phpdoc and the IdentityTranslator params similar to 90a20d7835
Commits
-------
7cdb260 [Translation] make IdentityTranslater consistent with normal translator
This PR was squashed before being merged into the master branch (closes#9540).
Discussion
----------
[Security] Added a missing field in SimpleAuthenticationHandler
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Added tests and made a small CS fix as a bonus.
Btw, the properties are protected. Do we really want to provide extension points here?
Commits
-------
d553347 [Security] Added a missing field in SimpleAuthenticationHandler
This PR was squashed before being merged into the master branch (closes#9536).
Discussion
----------
[FrameworkBundle] Update 2 dependencies (currently broken)
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| License | MIT
## Dependency on http-foundation
In b1a062d232 the `RequestStack` was moved to `http-foundation`, because of this the `framework-bundle` now also depends on `http-foundation`: b1a062d232 (diff-3) (at least in services.xml).
We got an error while installing the framework-bundle `2.4.*` with `prefer-stable` in our composer.json. We currently get framework-bundle 2.4 with http-kernel and http-foundation 2.3. An alternative fix would be raising the requirement of http-kernel to 2.4, but I think this is the "purest" declaration of dependencies.
Update 1
## Dependency on config
In 05e9ca7509 the config component was updated with an xml reference dumper for configuration. In the same commit the frameworkbundle dump command is updated to make use of the 2 newly introduced classes. It is currently possible to get an installation with config `2.3`, which in turn leads to an exception when running the dump-reference command:
```
$ app/console config:dump-reference framework
# Default configuration for extension with alias: "framework"
PHP Fatal error: Class 'Symfony\Component\Config\Definition\Dumper\YamlReferenceDumper
```
Update 2
## Dependency on http-kernel
We just found out that the dependency on http-kernel should also be updated to `2.4`. That's because the framework-bundle injects the `RequestStack` now and doesn't set the request on the `RouterListener` service anymore. Without updating the dependency it is possible to get a `RouterListener` that is not aware of the request stack (< 2.4) and none of your routes will work.
*tests are failing because of some race condition I guess?*
Commits
-------
fe5d8f8 [FrameworkBundle] Update 2 dependencies (currently broken)
This PR was merged into the master branch.
Discussion
----------
Add missing dependency to expression-language
Add missing dependency, without this dependency, PHPUnit can't run successfully.
Commits
-------
4206e98 Add missing dependency to expression-language
This PR was squashed before being merged into the master branch (closes#9529).
Discussion
----------
[ExpressionLanguage] Fixed conflict between punctation and range
The range operator (`..`) was actually lexed as 2 punctations (`.`). That causes any expression using the range to fail.
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Commits
-------
b521dc5 [ExpressionLanguage] Fixed conflict between punctation and range
My kernel has been moved and namespaced to `Cfs\Bundle\MultiSiteBundle\Kernel\CfsKernel`. This worked fine until a change was made to how the kernel temp stuff is handled in the warmup phase.
When the app generates its own cache (i.e you run cache without warmup and access the site) everything is generated ok and the .meta files generate the proper reference to the FQN of the Kernel.
However if the warmup is used, it uses `Cfs\Bundle\MultiSiteBundle\Kernel\CfsKerne_` as the temporary Kernel, and when it does "fix references to the Kernel in .meta files" it generates 2 errors.
1. It does not use a string safe tempKernel name, so it never finds the reference to the kernel
2. If you fix that, then it replaces the FQN of the tempKernel with `CfsKernel`, the non-namespaced name of the proper Kernel (it also leaves the character count wrong in the serialization `C:43:<class>` where 43 is the char count for the FQN above)
The two changes above fix this, by escaping the string and replacing it with a FQN Kernel Class name.
What are your thoughts on this?
This PR was merged into the master branch.
Discussion
----------
[Debug] Fixed a typo.
@pborreli style ;)
Commits
-------
df90c62 [Debug] Fixed a typo.
* 2.3: (25 commits)
bumped Symfony version to 2.2.11
updated VERSION for 2.2.10
update CONTRIBUTORS for 2.2.10
updated CHANGELOG for 2.2.10
fixed version
Request::overrideGlobals() may call invalid ini value
Force Luhn Validator to only work with strings
Fixed bug with lazy services
[Translation] fixed the error in the dumper test as described in #9475
deleted mixing string concatenation inside a sprintf
"__call()" should be displayed only if $this->magicCall is true
[Console] fix phpdoc and constructor default value
Add media-query for printing: Do not print the toolbar.
Fix ProgressHelper redraw when redrawFreq is greater than 1
Update toolbar.css.twig
slovenian translations fixed
slovenian translations fixed
[DependencyInjection] fixed YamlDumper did not make services private.
[FrameworkBundle] fix routing container parameter exception message
[Form] fix and unify phpdoc
...
Conflicts:
src/Symfony/Component/Console/Helper/ProgressHelper.php
src/Symfony/Component/DependencyInjection/ContainerBuilder.php
src/Symfony/Component/HttpKernel/Kernel.php
* 2.2:
bumped Symfony version to 2.2.11
updated VERSION for 2.2.10
update CONTRIBUTORS for 2.2.10
updated CHANGELOG for 2.2.10
fixed version
Request::overrideGlobals() may call invalid ini value
Force Luhn Validator to only work with strings
[Translation] fixed the error in the dumper test as described in #9475
[Console] fix phpdoc and constructor default value
Conflicts:
src/Symfony/Component/Console/Helper/ProgressHelper.php
src/Symfony/Component/HttpKernel/Kernel.php
This PR was submitted for the master branch but it was merged into the 2.2 branch instead (closes#9499).
Discussion
----------
Request::overrideGlobals() may call invalid ini value
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | perhaps
| Deprecations? | no
| Tests pass? | -
| Fixed tickets | -
| License | MIT
| Doc PR | -
According to http://php.net/manual/ja/ini.core.php ,
there's not variable_order, but variables_order (with trailing "s").
Perhaps it breaks BC for some developer who unsets (sets falsy value to )
'request_order' ini value and sets 'variable_order' manually?
Commits
-------
7f43fb8 Request::overrideGlobals() may call invalid ini value
According to http://php.net/manual/ja/ini.core.php ,
there's not variable_order, but variables_order (with trailing "s").
Perhaps it breaks BC for some developer who unsets
'request_order' ini value and sets 'variable_order' manually?
