If security isn't configured in the application, the `security.context` service, nor the `security.token_storage` service exists.
Therefore, if a third-party bundle relies on the app.user check in Twig templates, an exception was thrown about asking for an non-existing service.
Instead, this change check if the `security.context` actually exists before trying to use it, and return null otherwise.
Autoprepend appropriate blocks by the correct number of blank lines.
Handle most of the SymfonyStyle guide line breaks and gaps automatically and fix thing such as double blank lines between titles.
Add output tests.
Fix askQuestion method, which should not output anything when the input isn't interactive.
* 2.6:
bumped Symfony version to 2.6.10
updated VERSION for 2.6.9
updated CHANGELOG for 2.6.9
fixed tests
Avoid using the app global variable in the profiler templates
Conflicts:
src/Symfony/Component/HttpKernel/Kernel.php
- resolved possible PHP Fatal in \Symfony\Component\BrowserKit\Cookie::__toString
-resolved implicit magic methods calls
-resolved callable name case mismatches
This PR was merged into the 2.7 branch.
Discussion
----------
Adding UPGRADE-2.7 about small BC break in InputDefinition
Hi guys!
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #13220
| License | MIT
| Doc PR | n/a
In #13220, a small BC break was introduced when `InputDefinition::getSynopsis()` was given an optional argument. This, for example, breaks Behat 2.5, which overrides this method. I don't think this was noticed before, and it may be too late (or not worth it) to reverse the BC break.
So, I've at least doc'ed it.
Thanks!
Commits
-------
a7985d2 Fixing phpdoc typo
c35f2c8 Talking about getSynopsis()
This PR was merged into the 2.6 branch.
Discussion
----------
Avoid using the app global variable in the profiler templates
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | small one for people overwriting the WebDebugToolbarListener methods
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | https://github.com/silexphp/Silex-WebProfiler/issues/64
| License | MIT
| Doc PR | n/a
#14217 introduced a usage of the ``app`` global variables in profiler templates, while we previously removed all such usages to avoid the dependency on TwigBundle in these templates.
This keeps them usable outside the fullstack framework (for instance in Silex).
I had to do a small BC break to be able to pass the request in the place where we were not yet injecting the Request in the template.
Commits
-------
415c79e Avoid using the app global variable in the profiler templates
* 2.6: (21 commits)
Revert "bug #14262 [TwigBundle] Refresh twig paths when resources change. (aitboudad)"
InvalidResourceException file name
[Validators] Remove forgotten space in a translation key [nl]
[Validators] Correct translation key and content [nl]
bumped Symfony version to 2.6.9
updated VERSION for 2.6.8
updated CHANGELOG for 2.6.8
added missing CVE number
bumped Symfony version to 2.3.30
updated VERSION for 2.3.29
update CONTRIBUTORS for 2.3.29
updated CHANGELOG for 2.3.29
[Validators] Missing translations for arabic language.
Code style
fixed C
[HttpKernel][Bundle] Check extension implements ExtensionInterface
[DebugBundle] Fix config XSD
[CS] [Console] StreamOuput : fix loose comparison
[Framework][router commands] fixed failing test.
[HttpKernel] Do not call the FragmentListener if _controller is already defined
...
Conflicts:
src/Symfony/Component/HttpKernel/Kernel.php
src/Symfony/Component/Serializer/Normalizer/PropertyNormalizer.php
* 2.3:
Revert "bug #14262 [TwigBundle] Refresh twig paths when resources change. (aitboudad)"
InvalidResourceException file name
[Validators] Remove forgotten space in a translation key [nl]
[Validators] Correct translation key and content [nl]
added missing CVE number
bumped Symfony version to 2.3.30
updated VERSION for 2.3.29
update CONTRIBUTORS for 2.3.29
updated CHANGELOG for 2.3.29
[CS] [Console] StreamOuput : fix loose comparison
[DependencyInjection] Avoid unnecessary calls to strtolower()
Conflicts:
src/Symfony/Component/Console/Output/StreamOutput.php
src/Symfony/Component/HttpKernel/Kernel.php
This PR was merged into the 2.7 branch.
Discussion
----------
Adding notes about deprecated warnings in 2.7
| 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
Hi guys!
When I upgraded, my project exploded with deprecated warnings. I'm certain we'll get a lot of questions about these, so I think we should try to get ahead of these questions as much as possible. I'm adding this - if I'm mistaken in anything, please let me know. I also think we should mention this in the release blog post so that people are aware.
I'm about to propose a short docs entry to explain things there too (so that hopefully if someone googles for the issue, they'll find our explanation).
Thanks!
Commits
-------
e374274 Adding notes about deprecated warnings in 2.7
This PR was submitted for the 2.7 branch but it was merged into the 2.3 branch instead (closes#14762).
Discussion
----------
InvalidResourceException file name
Added filen ame to the InvalidResourceException,
Commits
-------
2e04e23 InvalidResourceException file name
This PR was merged into the 2.3 branch.
Discussion
----------
[Validators] Correct translation key and content [nl]
The nl file is not up to date. Correct changed translation key.
Commits
-------
5bc4085 [Validators] Correct translation key and content [nl]
This PR was merged into the 2.3 branch.
Discussion
----------
[Validators] Remove forgotten space in a translation key [nl]
The nl file is not up to date. Correct changed translation key.
Commits
-------
51e3c22 [Validators] Remove forgotten space in a translation key [nl]
This PR was merged into the 2.6 branch.
Discussion
----------
[Serializer] [2.6] Code style
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | none
| License | MIT
| Doc PR | none
This PR fixes some minor code style issues fabbot discovered while I was working on PR #14756.
Commits
-------
864136a Code style
This PR was merged into the 2.8 branch.
Discussion
----------
[Serializer] Support for array denormalization
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | none
| License | MIT
| Doc PR | none (yet)
This is a rebase of #14343 against the 2.8 branch.
I had to implement a couple of API endpoints that receive data as JSON-serialized arrays of objects. The current implementation of the Serializer component is capable of serializing such arrays, but so far this operation is a one-way-road. This PR is an attempt to change this.
## Demo
```php
class Dummy
{
public $foo;
public $bar;
public function __construct($foo = null, $bar = null)
{
$this->foo = $foo;
$this->bar = $bar;
}
}
$serializer = new \Symfony\Component\Serializer\Serializer(
array(
new \Symfony\Component\Serializer\Normalizer\PropertyNormalizer(),
new \Symfony\Component\Serializer\Normalizer\ArrayDenormalizer()
),
array(
new \Symfony\Component\Serializer\Encoder\JsonEncoder()
)
);
$json = $serializer->serialize(
array(
new Dummy('one', 'two'),
new Dummy('three', 'four')
),
'json'
);
echo $json . "\n\n";
// Deserialize the JSON string, so we get back to where we started from.
$data = $serializer->deserialize($json, 'Dummy[]', 'json');
var_dump($data);
```
By appending `[]` to the type parameter, you indicate that you expect to deserialize an array of objects of the given type. This is the same notation that phpDocumentor uses to indicate collections. The denormalization of the array is implemented recursively: The denormalizer simply calls `Serializer::denormalize()` on each element of the array. This way, the ArrayDenormalizer can be combined with any other denormalizer.
## Side effects
For this implementation, I had to touch `GetSetMethodNormalizer`, `PropertyNormalizer` and `CustomNormalizer`. Those classes expected `supportsDenormalization` to be called with a valid class in the `$type` parameter. Instead of throwing a reflection exception, they now simply return false. I'm not exactly sure, if this is should be considered to be a BC break.
This implementation violates the `SerializerInterface` which declared `deserialize()` to always return an object. But imho, the assumption that serialized data always represents an object is too restrictive anyway. Also, this declaration is not consistent with the `serialize()` method which accepts `mixed` as input type for the data to serialize.
## Other PRs
I've found an older PR adressing this issue, #12066. That PR was closed because the contributor did not reply to feedback.
Commits
-------
0573f28d Support for array denormalization.
* 2.3:
[Validators] Missing translations for arabic language.
fixed C
[HttpKernel][Bundle] Check extension implements ExtensionInterface
[HttpKernel] Do not call the FragmentListener if _controller is already defined
maxime.steinhausser
Abdellatif Ait boudad
Benjamin Laugueux
Christian Flothmann
Hassan Amouhzi
ogizanagi
Fabien Potencier
Vladimir Reznichenko
Ryan Weaver
Christophe Coevoet
Ladislav Tánczos
Kévin Dunglas
Nicolas Grekas