This PR was merged into the 3.3-dev branch.
Discussion
----------
[DependencyInjection] make the service container builder register its own self referencing definition
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | ~
| License | MIT
| Doc PR | ~
Commits
-------
9c97496b5f [DependencyInjection] make the service container builder register the definition of its related service container service (and aliases) in order to make compiler passes be able to reference the special service_container service.
This PR was merged into the 3.3-dev branch.
Discussion
----------
[Routing][DX] Add full route definition for invokable controller/class
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| License | MIT
| Doc PR | _not yet_
Currently the [`@Route`][1] annotation can be set on the class (for global parameters only). This PR allows you to define the full route annotation for _single_ controllers on the class.
Here a common use case of [ADR pattern][3] applied to Symfony:
**Before:**
```
use Sensio\Bundle\FrameworkExtraBundle\Configuration\Route;
/**
* @Route(service="AppBundle\Controller\Hello")
*/
class Hello
{
private $logger;
public function __construct(LoggerInterface $logger)
{
$this->logger = $logger;
}
/**
* @Route("/hello/{name}", name="hello")
*/
public function __invoke($name = 'World')
{
$this->logger->info('log entry...');
return new Response(sprintf('Hello %s!', $name));
}
}
```
**After:**
```
use Sensio\Bundle\FrameworkExtraBundle\Configuration\Route;
/**
* @Route("/hello/{name}", name="hello", service="AppBundle\Controller\Hello")
*/
class Hello
{
private $logger;
public function __construct(LoggerInterface $logger)
{
$this->logger = $logger;
}
public function __invoke($name = 'World')
{
$this->logger->info('log entry...');
return new Response(sprintf('Hello %s!', $name));
}
}
```
This feature does not break any behavior before and works under these conditions:
* The class cannot contain other methods with `@Route` annotation (otherwise, this works as before: used for global parameters).
* <del>The class `@Route` must have the `name` option defined (otherwise, the route is ignored).</del> This one is auto-generated if `null`.
* The class must be invokable: [`__invoke()` method][2] (otherwise, the route is ignored).
Btw, this PR fix the inconsistency with other route definitions (xml, yml) where the `_controller` parameter points to the class name only (i.e. without method).
[1]: https://github.com/symfony/symfony/tree/master/src/Symfony/Component/Routing/Annotation/Route.php
[2]: http://php.net/manual/en/language.oop5.magic.php#object.invoke
[3]: https://github.com/pmjones/adr
Commits
-------
34e360ade3 Add full route definition to invokable class
This PR was merged into the 3.3-dev branch.
Discussion
----------
[HttpKernel] Add a ContainerControllerResolver (psr-11)
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | N/A
| License | MIT
| Doc PR | N/A
Extracts the controller as service resolution from the framework bundle controller resolver in a `Symfony/Component/HttpKernel/Controller/Psr11ControllerResolver`, allowing you to use `HttpKernel` with your own psr-11 container.
Commits
-------
7bbae4159b [HttpKernel] Add a ContainerControllerResolver (psr-11)
This PR was merged into the 3.3-dev branch.
Discussion
----------
[Form] allow form types + form type extensions + form type guessers to be private services
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | ~
| License | MIT
| Doc PR | ~
This pull request is about making internal form services (aka form types, form type extensions and form type guessers) private. They used to be public until Symfony 3.2 for one valid reason: lazyness. However, Symfony 3.3 now comes with built-in mechanism to support effective lazy loading of private services with service locators and proxies.
This PR makes the `DependencyInjectionExtension` class of the `Form` component leverage these new DI component mechanisms. Form types, form type extensions and form type guessers can now be declared private as a best practice. We decided to make these services private as of Symfony 3.3 and of course it would break BC. But this PR introduces a BC layer using a Symfony trick to keep internal form services public. The service container currently has a known issue where private services are not really private if they're referenced by at least two other services in the container. We use this trick to maintain the legacy services public even though the new API relies on private ones. This trick is done thanks to the `deprecated.form.registry` and `deprecated.form.registry.csrf` fake services that will be removed in Symfony 4.0.
Commits
-------
600e75ce88 [Form] use new service locator in DependencyInjectionExtension class, so that form types can be made private at some point.
This PR was merged into the 3.3-dev branch.
Discussion
----------
[Config][FrameworkBundle] Lazy load resource checkers
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | N/A
| License | MIT
| Doc PR | N/A
`ResourceCheckerConfigCache::isFresh()` stops on the first resource checker supporting a resource (considered authoritative). Thus no need to instantiate other checkers.
Commits
-------
5a24ee2da1 [Config][FrameworkBundle] Lazy load resource checkers
This PR was squashed before being merged into the 3.3-dev branch (closes#21755).
Discussion
----------
[Routing] Optimised dumped router matcher, prevent unneeded function calls.
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | none
| License | MIT
| Doc PR | does not apply
The application I'm working on is fairly large. Because we had a routing issue (not caused by the framework) I looked through the dumped routing code. I spotted some easy wins. These changes brought down the time for the `match` method to run from ~ 7.5ms to ~2.5ms. It's not a lot, but it's something. I've profiled it several times with blackfire to confirm. The results were very consistent. Mind you, our application has quite a serious amount of routes, a little over 900.
Commits
-------
dd647ffc8a [Routing] Optimised dumped router matcher, prevent unneeded function calls.
This PR was merged into the 3.3-dev branch.
Discussion
----------
[FrameworkBundle][Config] Move ConfigCachePass from FrameworkBundle to Config
| Q | A
| ------------- | ---
| Branch? | master<!--see comment below-->
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | yes
| Tests pass? | yes/no
| Fixed tickets | - <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR | <!--highly recommended for new features-->
This MR is part of the #21284 todo list
<!--
- Bug fixes must be submitted against the lowest branch where they apply
(lowest branches are regularly merged to upper ones so they get the fixes too).
- Features and deprecations must be submitted against the master branch.
- Please fill in this template according to the PR you're about to submit.
- Replace this comment by a description of what your PR is solving.
-->
Commits
-------
bce445f452 Move ConfigCachePass from FrameworkBundle to Config
This PR was merged into the 3.3-dev branch.
Discussion
----------
[DI] Simplify AutowirePass and other master-only cleanups
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
A few minor cleanups and fixes, and an overall simplification of AutowirePass.
Commits
-------
34e5cc7698 [DI] Simplify AutowirePass and other master-only cleanups
This PR was merged into the 2.7 branch.
Discussion
----------
[Serializer] Reduce nesting in YamlFileLoader
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
We don't need to check if $this->classes is empty, because isset takes care of it in next call anyway
Diffs on GH are hard to read for this type of change, here is old and new code:
```php
public function loadClassMetadata(ClassMetadataInterface $classMetadata)
{
if (null === $this->classes) {
$this->classes = $this->getClassesFromYaml();
}
if (!$this->classes) {
return false;
}
if (isset($this->classes[$classMetadata->getName()])) {
$yaml = $this->classes[$classMetadata->getName()];
if (isset($yaml['attributes']) && is_array($yaml['attributes'])) {
...
}
return true;
}
return false;
}
```
```php
public function loadClassMetadata(ClassMetadataInterface $classMetadata)
{
if (null === $this->classes) {
$this->classes = $this->getClassesFromYaml();
}
if (!isset($this->classes[$classMetadata->getName()])) {
return false;
}
$yaml = $this->classes[$classMetadata->getName()];
if (isset($yaml['attributes']) && is_array($yaml['attributes'])) {
...
}
return true;
}
```
Commits
-------
45f0b16 [Serializer] Reduce nesting in YamlFileLoader
This PR was merged into the 3.3-dev branch.
Discussion
----------
Revert "feature #21792 [Security] deprecate multiple providers in context listener (xabbuh)"
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #21791
| License | MIT
| Doc PR |
#21792 was a mistake as pointed out by @slaci (see https://github.com/symfony/symfony/pull/21791#issuecomment-282990746) and @stof (see https://github.com/symfony/symfony/pull/21792#issuecomment-282980046).
Commits
-------
3cfa0c7 Revert "feature #21792 [Security] deprecate multiple providers in context listener (xabbuh)"
This PR was squashed before being merged into the 3.3-dev branch (closes#21792).
Discussion
----------
[Security] deprecate multiple providers in context listener
Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | yes
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
Passing multiple user providers to the context listener does not make
much sense. The listener is only responsible to refresh users for a
particular firewall. Thus, it must only be aware of the user provider
for this particular firewall.
Commits
-------
53df0de7fc [Security] deprecate multiple providers in context listener
fbd9f88e31 [SecurityBundle] only pass relevant user provider
* 3.2:
[SecurityBundle] only pass relevant user provider
[Intl] Make tests pass after the ICU data update
[Intl] Update ICU data to 58.2
do not register the test listener twice
[DependencyInjection] removed dead code.
[Yaml] Stop replacing NULLs when merging
[WebServerBundle] fixed html attribute escape
* 2.8:
[SecurityBundle] only pass relevant user provider
[Intl] Make tests pass after the ICU data update
[Intl] Update ICU data to 58.2
do not register the test listener twice
[DependencyInjection] removed dead code.
[Yaml] Stop replacing NULLs when merging
[WebServerBundle] fixed html attribute escape
* 2.7:
[SecurityBundle] only pass relevant user provider
[Intl] Make tests pass after the ICU data update
[Intl] Update ICU data to 58.2
[DependencyInjection] removed dead code.
[Yaml] Stop replacing NULLs when merging
This PR was merged into the 3.3-dev branch.
Discussion
----------
Remove some container injections in favor of service locators
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | https://github.com/symfony/symfony/pull/21553#issuecomment-279214666
| License | MIT
| Doc PR | n/a
Commits
-------
8293b753cf Replace some container injections by service locators
0be9ea8ba1 [EventDispatcher] Fix abstract event subscribers registration
This PR was merged into the 3.3-dev branch.
Discussion
----------
[VarDumper] Allow seamless use of Data clones
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | yes
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
By implementing `ArrayAccess`, `Countable`, `IteratorAggregate`, `__get`, `__isset` and `__toString`, VarDumper's `Data` objects become seamless and behave almost identically from their original clones values, especially from the PoV of Twig.
In data collectors, this allows replacing the many nested calls to `cloneVar` by a single one.
This makes the code simpler, and should make a significant difference in term of performance.
Todo:
- [x] push a Blackfire profile comparison
- [x] double check that the profiler works as expected.
Commits
-------
ab716c64de [VarDumper] Allow seamless use of Data clones
This PR was merged into the 3.3-dev branch.
Discussion
----------
[DI] Always consider abstract getters as autowiring candidates
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes (a missing part of getter autowiring really)
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
When a definition is set to be autowired with no method explicitly configured, we already wire the constructor.
We should also autowire abstract getters - with the same reasoning that makes us autowire the constructor: without concrete getters, the class is unusable. This just makes it usable again.
Commits
-------
8f246bde1d [DI] Always consider abstract getters as autowiring candidates
Passing multiple user providers to the context listener does not make
much sense. The listener is only responsible to refresh users for a
particular firewall. Thus, it must only be aware of the user provider
for this particular firewall.
This PR was squashed before being merged into the 2.7 branch (closes#21790).
Discussion
----------
[Intl] Update ICU data to 58.2
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
ICU 58.2 was released: http://site.icu-project.org/download/58
Commits
-------
d135e5cd2f [Intl] Make tests pass after the ICU data update
65faa1043d [Intl] Update ICU data to 58.2
This PR was merged into the 3.3-dev branch.
Discussion
----------
[DI] Fix ordering of tags inheritance
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Tag inheritance should have predictable ordering, from child to parents.
This PR also adds tests for ResolveDefinitionInheritancePass (heavily inspired by ResolveDefinitionTemplatePass).
Commits
-------
18e7681fc5 [DI] Fix ordering of tags inheritance
