This PR was merged into the 2.1 branch.
Commits
-------
3553276 [ConfigDumpReference] avoid notice for variable nodes
Discussion
----------
[ConfigDumpReference] avoid notice for variable nodes
When a variable node has an array as default value, a notice occurs later on because of an "array to string conversion", which is turned to an exception in debug mode (mandatory in order to run this command).
* 2.1: (28 commits)
Delete use of CreationExeption
[Form] Fixed error message in PropertyPath to not advice to use a non-existing feature
[Form] Fixed creation of multiple money fields with different currencies
[Form] Fixed setting the "data" option to an object in "choice" and "entity" type
Fixed Serbian plural translations.
Fixed IPv6 Check in RequestMatcher
Fix typo
change what I think is a typo
[Console] Fix error when mode is not in PATH
[WebProfilerBundle] fixed macro usage (to be forward compatible with Twig 2.x)
Change monolog require-dev to use the branch alias instead of dev-master
[FrameworkBundle] partially reverted previous merge
[2.1] Added missing error return codes in commands
Made the router lazy when setting the context
[WebProfilerBundle] fixed typos
Fix incorrect variable in FileProfilerStorage
UnitTest fix
UnitTest fix
added a unit test
fixed#5384
...
This PR was squashed before being merged into the 2.1 branch (closes#5586).
Commits
-------
6b66bc3 [2.1] Added missing error return codes in commands
Discussion
----------
[2.1] Added missing error return codes in commands
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
License of the code: MIT
See: #5585
---------------------------------------------------------------------------
by fabpot at 2012-09-24T12:10:47Z
Exit code values are standardized and some values have some well-defined meaning. Have a look here for more info: https://github.com/symfony/symfony/blob/master/src/Symfony/Component/Process/Process.php#L67
* 2.1:
fixed CS
added doc comments
added doc comments
[Validator] Updated swedish translation
Update src/Symfony/Component/Validator/Resources/translations/validators.de.xlf
[2.1] Exclude tests from zips via gitattributes
[HttpKernel][Translator] Fixed type-hints
Updated lithuanian validation translation
[DomCrawler] Allows using multiselect through Form::setValues().
[Translation] forced the catalogue to be regenerated when a resource is added (closes symfony/Translation#1)
Unit test for patched method OptionsResolver::validateOptionValues().
validateOptionValues throw a notice if an allowed value is set and the corresponding option isn't.
[Form] Hardened code of ViolationMapper against errors
[HttpFoundation] Fixed#5611 - Request::splitHttpAcceptHeader incorrect result order.
[Form] Fixed negative index access in PropertyPathBuilder
Update src/Symfony/Component/Validator/Resources/translations/validators.ro.xlf
Conflicts:
src/Symfony/Component/DomCrawler/Form.php
src/Symfony/Component/Process/Process.php
This PR was merged into the master branch.
Commits
-------
4b86765 [FrameworkBundle] recursively resolve container parameter placeholders for arrays in router _defaults
Discussion
----------
[2.2] [FrameworkBundle] avoid trying to resolve container placeholders on arrays on router _defaults
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
Fixes the following tickets: ~
Todo: ~
License of the code: MIT
Documentation PR: ~
Permits to pass arrays in route `_defaults`.
---------------------------------------------------------------------------
by stof at 2012-07-20T13:07:36Z
This seems weird. An array could contain parameters in it.
---------------------------------------------------------------------------
by docteurklein at 2012-07-20T13:17:00Z
@stof An object too then, no ? Why accepting objects but not arrays ? Would you propose to recursively resolve array values ?
---------------------------------------------------------------------------
by stof at 2012-07-20T13:31:06Z
@docteurklein Resolving array values recursively would be consistent with the way the DIC parameters are resolved. I don't really see how you would resolve objects (and btw, it is pretty much an edge case as you cannot really put an object in your routes if you define them in your YAML or XML config files or with annotations)
---------------------------------------------------------------------------
by docteurklein at 2012-07-20T13:36:43Z
@stof I agree. I can manage recursive array resolving if needed.
---------------------------------------------------------------------------
by fabpot at 2012-07-23T13:58:07Z
Can you squash your commits before I merge? Thanks.
---------------------------------------------------------------------------
by docteurklein at 2012-07-23T14:39:17Z
@fabpot done.
The ContainerAwareTraceableEventDispatcher class was tied to both the
Symfony container and the HttpKernel profiler. It made it non reusable
in another context.
The new TraceableEventDispatcher only keeps the HttpKernel profiler
integration and is able to wrap any other event dispatcher. It makes it
reusable in frameworks using the Symfony HttpKernel component like
Silex.
The only drawback is that we don't have access to the listener
priorities in the collected data anymore (but the listeners are still
ordered correctly). The change is still worth it I think.
Commits
-------
22e9036 updated CHANGELOG
bafe890 [FrameworkBundle] changed Client::enableProfiler() behavior to fail silently when the profiler is not available (it makes it easier to write functional tests)
f41872b [FrameworkBundle] added a way to enable the profiler for the very next request in functional tests (closes#4307)
67b91e5 [HttpKernel] added a way to enable a disable Profiler
Discussion
----------
[2.2] added a way to enable the profiler for the very next request in a functional test
Bug fix: yes/no
Feature addition: yes
Backwards compatibility break: no
Symfony2 tests pass: yes
Fixes the following tickets: #4307
Todo: -
License of the code: MIT
Documentation PR: should be done before merging
After merging this PR, we need to disable the profiler in the test environment in Symfony SE.
Commits
-------
3f8127c fixed '0' problem
7bec460 fixed phpdoc
4c5bfab [FrameworkBundle] non-permanent redirect should be status code 404 according to spec
Discussion
----------
[FrameworkBundle] non-permanent redirect to unknown location with 404
according to spec: http://www.w3.org/Protocols/rfc2616/rfc2616-sec10.html see 410 Gone
bc break: tiny when omitting 2 parameter (I can avoid this with `func_num_args` but i think its not necessary and makes the code strange and inconsistent)
* 2.0:
updated VERSION for 2.0.17
updated CHANGELOG for 2.0.17
updated vendors for 2.0.17
fixed XML decoding attack vector through external entities
prevents injection of malicious doc types
disabled network access when loading XML documents
refined previous commit
prevents injection of malicious doc types
standardized the way we handle XML errors
Redirects are now absolute
Conflicts:
CHANGELOG-2.0.md
src/Symfony/Component/DependencyInjection/Loader/XmlFileLoader.php
src/Symfony/Component/DomCrawler/Crawler.php
src/Symfony/Component/HttpKernel/Kernel.php
tests/Symfony/Tests/Component/DependencyInjection/Loader/XmlFileLoaderTest.php
tests/Symfony/Tests/Component/Routing/Loader/XmlFileLoaderTest.php
tests/Symfony/Tests/Component/Serializer/Encoder/XmlEncoderTest.php
tests/Symfony/Tests/Component/Translation/Loader/XliffFileLoaderTest.php
tests/Symfony/Tests/Component/Validator/Mapping/Loader/XmlFileLoaderTest.php
vendors.php
Commits
-------
85a53c1 [FrameworkBundle] fixed *FrameworkExtensionTest::testTranslator fail on Windows on master branch
Discussion
----------
[FrameworkBundle] fixed *FrameworkExtensionTest::testTranslator fail on Windows on master branch
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
Fixes the following tickets: #5345
Todo: -
License of the code: MIT
Documentation PR:
fixed using some
str_replace('/', DIRECTORY_SEPARATOR, <value>)
so it works on Windows because it replaces the / in \ and shouldn't affect linux based OS, since it will replace / with /.
Not very nice, but it works, if anyone wants to do better, he/she is most welcome.
Commits
-------
933e821 Add minimum-stability (dev) in each component
Discussion
----------
Add minimum-stability (dev) in each component
This fixes the ability to run the test suite in each component if a `composer install` is needed.
---------------------------------------------------------------------------
by stof at 2012-08-22T13:57:14Z
If you really want to run the testsuite standalone, some dev requirements are missing (SecurityBundle needs the FrameworkBundle for its functional tests for instance). If you have some time to check the missing dev requirement, it would be great.
Anyway, 👍 for this
---------------------------------------------------------------------------
by willdurand at 2012-08-22T13:59:15Z
Yes I already did that once. I'll try to fix more components later.
On Wed, Aug 22, 2012 at 3:57 PM, Christophe Coevoet <
notifications@github.com> wrote:
> If you really want to run the testsuite standalone, some dev requirements
> are missing (SecurityBundle needs the FrameworkBundle for its functional
> tests for instance). If you have some time to check the missing dev
> requirement, it would be great.
> Anyway, [image: 👍] for this
>
> —
> Reply to this email directly or view it on GitHub<https://github.com/symfony/symfony/pull/5318#issuecomment-7934886>.
>
>
---------------------------------------------------------------------------
by stof at 2012-08-22T14:02:23Z
Well, I think most components should be good now (as some work has been done on them). But the bridges and bundles may need some work (bundles were not having any dev requirements until yesterday when @guilhermeblanco added some on FrameworkBundle)
---------------------------------------------------------------------------
by pborreli at 2012-08-22T14:14:00Z
what about having for each READ-ONLY repo his own .travis.yml and travisci hook activated ?
---------------------------------------------------------------------------
by fabpot at 2012-08-22T14:30:13Z
please, don't add more travis files. The main already tests everything, and that's all we need.
---------------------------------------------------------------------------
by stof at 2012-08-22T14:33:46Z
@pborreli tests should not be different for subtree split repos as the code is the same and the tests are the same (except that more tests could be skipped because of missing deps).
Note that for the bundles, it is likely to be different currently as I think some skip tests are missing (just like dev requirements are). But fixing this does not require enablign travis.
---------------------------------------------------------------------------
by pborreli at 2012-08-22T14:42:30Z
ok, i was just thinking about a way to be sure each component is usable individually but yeah that would require to relaunch each tests and add a bunch of travis files + hook
---------------------------------------------------------------------------
by hason at 2012-08-24T13:12:04Z
@stof, @eriksencosta, @fabpot: Tests are different for Locale component, see #5235
---------------------------------------------------------------------------
by stof at 2012-08-24T13:35:07Z
@hason no. You also need to do it when running the tests of the Locale component as part of the full run.
Commits
-------
9c20634 fixes pre for var_dump with xdebug
Discussion
----------
Displaying var_dump with xdebug in exceptions
When debugging code I often use `var_dump` to quickly look into variables. Since 2.1 alle output generated by `var_dump` is displayed in one line. http://screencast.com/t/11LuIlIdHsvP
It seems to be no problem for small objects, but it becomes a real pain when displaying huge arrays or objects.
This is caused by the changed word-wrapping for the pre tag introduced in #3827
With fix: http://screencast.com/t/GdA3dkpWxU
---------------------------------------------------------------------------
by dlsniper at 2012-08-17T17:22:38Z
👍
Commits
-------
cdfbe72 handle inheritance in config:dump-reference when a bundle name is passed to the command
Discussion
----------
handle inheritance in config:dump-reference
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: ?
Fixes the following tickets: -
License of the code: MIT
Currently when passing in a bundle name that identifies a bundle that is a parent to another bundle, it will not return the reference for the requested bundle, but for the inheriting (even if that bundle has no configuration class).
F.e.
app/console config:dump-reference SymfonyCmfBlockBundle
will fail if there is a Bundle SandboxBlockBundle that has SymfonyCmfBlockBundle set as the parent.
* 2.0:
Fixes incorrect class used in src/Symfony/Bundle/FrameworkBundle/Console/Application.php
[FrameworkBundle] added test for fix broken command registration
corrected phpdoc
Issue must be related to commit 7a5f614240 (merged 2.0), specifically this file src/Symfony/Bundle/FrameworkBundle/Console/Application.php, lines 86-88.
Presumably to do "instanceof Bundle" correct class has to be imported at the top of the file:
instead of
use Symfony\Component\HttpKernel\Bundle;
this should be
use Symfony\Component\HttpKernel\Bundle\Bundle;
Conflicts:
src/Symfony/Bundle/FrameworkBundle/Console/Application.php
Commits
-------
0b78fdf Only call registerCommand on bundles that is an instance of Bundle
Discussion
----------
Only call registerCommand on bundles that is an instance of Bundle
Fixes GH-5133
---------------------------------------------------------------------------
by travisbot at 2012-08-01T09:41:05Z
This pull request [fails](http://travis-ci.org/symfony/symfony/builds/2008252) (merged 0b78fdff into 1da896dc).
---------------------------------------------------------------------------
by henrikbjorn at 2012-08-01T10:05:00Z
Build failed because of HTTP request error.
---------------------------------------------------------------------------
by lsmith77 at 2012-08-01T11:31:08Z
wondering if it would be good if you could include the commit from #5133 in this PR .. then we get the test and the fix at once.
Commits
-------
b982883 [Form] Moved FormHelper back to FrameworkBundle
cb62d05 [Form] [Validator] Fixed issues mentioned in the PR
2185ca8 [Validator] Added entry point "Validation" for more convenient usage outside of Symfony2
ed87361 [Form] Moved FormHelper creation to TemplatingExtension
87ccb6a [Form] Added entry point "Forms" for more convenient usage outside of Symfony
Discussion
----------
[Form] [Validator] Added more convenient entry points for stand-alone usage
Bug fix: no
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
Fixes the following tickets: -
Todo: -
This PR greatly simplifies the usage of the Form and Validator component when used outside of Symfony2. Check out the below code to get an idea about the simplified usage:
```php
<?php
use Symfony\Component\Validator\Validation;
use Symfony\Component\Validator\Mapping\Cache\ApcCache;
use Symfony\Component\Form\Forms;
use Symfony\Component\Form\Extension\HttpFoundation\HttpFoundationExtension;
use Symfony\Component\Form\Extension\Csrf\CsrfExtension;
use Symfony\Component\Form\Extension\Csrf\CsrfProvider\SessionCsrfProvider;
use Symfony\Component\Form\Extension\Templating\TemplatingExtension;
use Symfony\Component\Form\Extension\Validator\ValidatorExtension;
use Symfony\Component\HttpFoundation\Session;
use Symfony\Component\Templating\PhpEngine;
$session = new Session();
$secret = 'V8a5Z97e...';
$csrfProvider = new SessionCsrfProvider($session, $secret);
$engine = new PhpEngine(/* ... snap ... */);
$validator = Validation::createValidator();
// or
$validator = Validation::createValidatorBuilder()
->addXmlMapping('path/to/mapping.xml')
->addYamlMapping('path/to/mapping.yml')
->addMethodMapping('loadValidatorMetadata')
->enableAnnotationMapping()
->setMetadataCache(new ApcCache())
->getValidator();
$formFactory = Forms::createFormFactory();
// or
$formFactory = Forms::createFormFactoryBuilder()
// custom types, if you're too lazy to create an extension :)
->addType(new PersonType())
->addType(new PhoneNumberType())
->addTypeExtension(new FormTypeHelpTextExtension())
// desired extensions (CoreExtension is loaded by default)
->addExtension(new HttpFoundationExtension())
->addExtension(new CsrfExtension($csrfProvider))
->addExtension(new TemplatingExtension($engine, $csrfProvider, array(
'FormBundle:Form'
))
->addExtension(new ValidatorExtension($validator))
->getFormFactory();
$form = $formFactory->createBuilder()
->add('firstName', 'text')
->add('lastName', 'text')
->add('age', 'integer')
->add('gender', 'choice', array(
'choices' => array('m' => 'Male', 'f' => 'Female'),
))
->getForm();
if (isset($_POST[$form->getName()])) {
$form->bind($_POST[$form->getName()]);
if ($form->isValid()) {
// do stuff
}
}
return $engine->render('AcmeHelloBundle:Hello:index.html.php', array(
'form' => $form->createView(),
));
```
---------------------------------------------------------------------------
by bschussek at 2012-07-30T10:08:42Z
I should maybe add a comment about the benefits of this change, in case they are not self-explanatory:
* class construction with default configuration is now a one-liner
* userland code is decoupled from core implementations → userland code doesn't break if we change constructor signatures
* easier to understand, since many core classes are now created internally
* easy to discover the possible settings → just look at (FormFactory|Validator)BuilderInterface
* usage of custom interface implementations is supported, just like before
---------------------------------------------------------------------------
by fabpot at 2012-07-31T08:18:53Z
The new syntax is great.
I have one comment though about this PR about support of PHP as a templating system (support for Twig is provided by the bridge and it was already easy to configure Twig as a templating system for forms -- see Silex for instance).
The `FormHelper` has been moved into the Form component. This helper is only useful when using the PHP templating system (which is not what we recommend people to use), but the default templates are still in the Framework bundle. So using the Form component as standalone with PHP as a templating system still requires to install the bundle to get access to the default templates. Am I missing something? Do we want to move the PHP templates to the Form component too?
---------------------------------------------------------------------------
by stof at 2012-07-31T08:28:28Z
@fabpot it is even worse than that: the FormHelper currently uses the theme by using ``$theme . ':' . $block . '.html.php`` IIRC. This is not compatible with the default template name parser of the component expecting a path. And the FrameworkBundle template name parser does not support accessing a template outside a bundle AFAIK.
So moving the templating to the component would require some refactoring in the FormHelper and the template name parser. However, I think it is worth it. Some people complained that using the form rendering (outside the full-stack framework) was requiring either setting up Twig with the bridge, or adding FrameworkBundle in the project (which means including most of the code of the full-stack framework). Having the Templating rendering in the standalone component could be a great idea
---------------------------------------------------------------------------
by fabpot at 2012-07-31T08:42:53Z
But then, I don't want to promote the Templating component or the PHP templating system. Twig is always a better alternative and this should be what people use most of the time, PHP being the rare exception.
Anyway, we are too close from the first 2.1 RC, so any big refactoring will have to wait for 2.2.
---------------------------------------------------------------------------
by stof at 2012-07-31T09:02:10Z
then maybe we should keep the FormHelper in FrameworkBundle for now as it is tied to the FrameworkBundle template name parser anyway currently.
---------------------------------------------------------------------------
by bschussek at 2012-07-31T14:22:35Z
> it it is even worse than that: the FormHelper currently uses the theme by using ``$theme . ':' . $block . '.html.php`` IIRC. This is not compatible with the default template name parser of the component expecting a path.
This is why the templates are still in FrameworkBundle. I think they should be moved too, but then we have to change
* the default theme to an absolute file path
* the FrameworkBundle name parser to accept absolute paths
I think this can wait until 2.2. Baby steps.
> I don't want to promote the Templating component or the PHP templating system.
We can both promote Twig while making Templating as easy to use as possible. If people want to use Templating, they probably have a reason. We don't have to make their lives more painful than necessary.
Btw: Templating is a *lot* faster for rendering forms than Twig. On Denis' form, Templating takes 1.15 seconds while Twig takes 2.
About moving the helpers, we have two choices:
* Move each helper to the respective component. This would not require new releases of the Templating component when we add more helpers in other component.
* Move all helpers to Templating. This does not make that much sense for Form, as then Form has support for Templating (TemplatingRendererEngine) and Templating has support for Form (FormHelper), which is a bit weird. I personally prefer a stacked architecture, where Templating is at the bottom and Form-agnostic, and Form (or any other component) builds upon that.
I'm fine with both approaches. I'll move FormHelper back to FrameworkBundle, and we can decide for a direction in 2.2.
---------------------------------------------------------------------------
by bschussek at 2012-07-31T14:36:30Z
Done.
Commits
-------
03bbaaf [Routing] Add an interface for configuring strict_parameters
Discussion
----------
[RFC][Routing] Add an interface for configuring strict_parameters
This is a proposal to fix#4697 (related to #4592).
The main point left to discuss was the name of the interface, which is now `LenientInterface`. We could change the name to anything else is someone has a better idea.
@stof @Tobion what do you think ?
---------------------------------------------------------------------------
by stof at 2012-07-30T16:34:20Z
@vicb I already said I had no idea to name it, and it has not changed. :)
So let's wait for other people to see if they have a better idea
---------------------------------------------------------------------------
by breerly at 2012-07-30T16:38:38Z
Maybe `PermissibleInterface` or `PermissiveInterface`.
---------------------------------------------------------------------------
by Partugal at 2012-07-30T17:00:09Z
`StrictUrlGeneratorInterface`, `StrictParametersInterface` or `StrictInterface`
---------------------------------------------------------------------------
by pborreli at 2012-07-30T17:04:46Z
👍 for `PermissiveInterface`
---------------------------------------------------------------------------
by stof at 2012-07-30T17:07:59Z
yes, because the Router currently can only use this interface to set it to ``not-strict``. It assumes that the url generator is already strict by default (which is probably a bad assumption btw as the base class for the generated generator can be changed)
---------------------------------------------------------------------------
by pborreli at 2012-07-30T17:09:33Z
@stof thx, got it
---------------------------------------------------------------------------
by Partugal at 2012-07-30T17:10:03Z
this interface realize setting Strict by setStrictParameters, and get by getStrictParameters, and imho named it by `Strictable` is more logic
---------------------------------------------------------------------------
by pborreli at 2012-07-30T17:11:07Z
@Partugal let's try to find an english term :)
---------------------------------------------------------------------------
by Partugal at 2012-07-30T17:11:31Z
)
---------------------------------------------------------------------------
by breerly at 2012-07-30T17:15:23Z
@Partugal I like using "able" in interface names because it describes a behavior instead of a noun. This type of naming makes following the Interface Segregation Principle easy to follow. Good work.
---------------------------------------------------------------------------
by vicb at 2012-07-30T18:24:26Z
As explained by @stof I did not consider `StrictInterface` because as of now the interface is used to disabled the strict bevahior (which is enabled by default).
I am not satisfied with `PermissiveInterface` / `LenientInterface` because implementing this interface does not mean that the generator will be permissive but only that the behavior is configurable - yes I did consider `Configurable` but the term is a too vague.
---------------------------------------------------------------------------
by breerly at 2012-07-30T18:35:45Z
I see. Perhaps ```StrictConfigurableInterface``` would do the trick.
---------------------------------------------------------------------------
by Tobion at 2012-07-30T21:02:21Z
I think renaming strict_parameters to `strict_requirements` is the way to go because it determines how requirements are handled when generating a URL. Also we should allow another option:
strict_requirements = true: throw exception for mismatching requirements
strict_requirements = null: return null as URL for mismatching requirements and log it.
strict_requirements = false: return the URL with the given parameters without checking the requirements and don't log it.
(Maybe use constants for these).
The Interface I would then call `ConfigurableRequirementsInterface` or `RequirementsHandlingInterface`.
---------------------------------------------------------------------------
by vicb at 2012-07-31T07:23:24Z
Thanks all for the feeback, this is what is now implemented:
- A `ConfigurableRequirementsInterface` that should be implemented by generators that can be configure not to throw an exception when the parameters do not match the requirements,
- The interface has two methods: `setStrictRequirements()` and `isStrictRequirements()`,
- `setStrictRequirements()` always gets called to configure the generator (whatever the option value is)
Note: The Router option name has not changed (i.e. `strict_parameters`)
Does that fit everyone ?
---------------------------------------------------------------------------
by vicb at 2012-07-31T07:39:22Z
So the option name is now consistent (`strict_requirements`) with the interface. We should sync the change [in the standard edition](https://github.com/symfony/symfony-standard/blob/master/app/config/config.yml#L11) if we agree to merge this.
---------------------------------------------------------------------------
by fabpot at 2012-07-31T07:51:47Z
@vicb you forgot to rename the property in `UrlGenerator` as @stof mentioned above.
---------------------------------------------------------------------------
by vicb at 2012-07-31T07:59:57Z
@fabpot fixed. If the code is ok, I'll squash the commits and open a PR on symfony-standard
Commits
-------
1474aa5 [Form] Fixed consideration of Twig's template inheritance and added another performance-improving check
b4ec7f5 Fixed my rubbish English
d11f8b5 [Form] Fixed passing of variables in the FormRenderer
629093e [Form] Extracted common parts of FormHelper and FormExtension into separate classes
216c539 [Form] Implemented a more intelligent caching strategy in FormHelper (PHP +100ms, Twig +100ms)
Discussion
----------
[Form] Merged FormHelper and FormExtension and implemented a better caching strategy
Bug fix: no
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
Fixes the following tickets: -
Todo: -
This PR extracts common parts of `FormHelper` and `FormExtension` into implementations of the new interfaces `FormRendererInterface` and `FormRendererEngineInterface`. The implemented `AbstractRendererEngine` features a more intelligent caching strategy than the one used before. When this strategy was implemented directly in `FormHelper`, the performance of [this specific, heavy form](http://advancedform.gpserver.dk/app_dev.php/taxclasses/1) could be improved from **2.5** to **2.25 seconds** on my machine for PHP templates.
Due to the abstraction and delegation, the performance gain is not that big anymore, but we still have a performance gain of about **0.1 seconds** for both PHP and Twig in the above example. The second, big improvement of this PR is maintainability - the differences between PHP and Twig templates are now contained in relatively small classes - and extendability (it is very easy now to support different template engines).
---------------------------------------------------------------------------
by stof at 2012-07-14T13:47:19Z
should a similar refactoring be done for the [Twig rendering](https://github.com/symfony/symfony/blob/master/src/Symfony/Bridge/Twig/Extension/FormExtension.php) ?
---------------------------------------------------------------------------
by bschussek at 2012-07-14T13:49:25Z
Yes. I would like to merge the common parts of Twig's FormExtension and PHP's FormHelper into an abstract class. Before that I need to have a [working, heavy Twig Form](https://twitter.com/webmozart/status/224135287377371138) in order to measure whether I don't actually decrease the performance with Twig. Can you help me there?
---------------------------------------------------------------------------
by vicb at 2012-07-16T21:48:24Z
Would it make sense to create a 'renderer' folder in the form component and move related classes there ?
---------------------------------------------------------------------------
by stof at 2012-07-16T22:06:58Z
@vicb It makes sense to keep the Twig renderer in the brisge. This is what the bridge is about. Moving the Twig class to the component would not be consistent. And the PHP renderer is already in the component (but it could make sense to move the helper from FrameworkBundle to the TemplatingExtension of the Form component though)
---------------------------------------------------------------------------
by vicb at 2012-07-16T22:16:50Z
@stof I was only referring to the classes located in the Component/Form folder.
---------------------------------------------------------------------------
by vicb at 2012-07-16T22:27:27Z
Overall I don't really know what to think of this PR. PHP and Twig use a different way to support blocks:
- PHP has one block per file,
- Twig could have many blocks per templates.
I am not sure if this PR is optimal for Twig and improves maintainability ?
---------------------------------------------------------------------------
by stof at 2012-07-16T22:46:11Z
@vicb it avoids duplicating the whole rendering logic for each engine (there is at least a third one in [SmartyBundle](https://github.com/noiselabs/SmartyBundle/blob/master/Extension/FormExtension.php) btw)
---------------------------------------------------------------------------
by bschussek at 2012-07-17T07:16:42Z
@vicb I don't think a renderer subfolder makes sense. The interfaces belong to the main namespace, and then the subfolder would only contain two classes.
Considering maintainability for Twig, I think that this PR in fact increases it. TwigExtension before always had to check the whole type hierarchy, while now the code in AbstractRendererEngine makes sure that this process is speeded up.
Before:
```
load _some_entity_field_label:
- check _some_entity_field_label
- check entity_label
- check choice_label
- check form_label
load _some_other_entity_field_label
- check _some_other_entity_field_label
- check entity_label
- check choice_label
- check form_label
a.s.o.
```
After:
```
load _some_entity_field_label:
- check _some_entity_field_label
- check entity_label (hits the cache if entity_label was checked before)
- check choice_label (hits the cache if choice_label was checked before)
- check form_label
load _some_other_entity_field_label
- check _some_other_entity_field_label
- check entity_label (now definitely hits the cache)
a.s.o.
```
Since many fields share the same ancestors in the inheritance tree, this definitely improves performance.
As can also be deducted here, custom block names such as `_some_entity_field_label` are now a major drawback. There is nothing we can cache for them, so they need to be checked for every individual block that we load. Removing this feature surprisingly gains no performance for Twig (I need to investigate why at some point), but it speeds up rendering for **250ms** using the PHP engine on [this example form](advancedform.gpserver.dk/app_dev.php/taxclasses/1), dropping the rendering time from 1.25 to 1 sec on my local machine. I'm not sure what we should do here.
---------------------------------------------------------------------------
by stof at 2012-07-17T07:21:31Z
@bschussek could it be possible to have an implementation checking the custom block and another one skipping it ? This way, the user could disable this feature when he does not need it.
---------------------------------------------------------------------------
by bschussek at 2012-07-17T07:38:34Z
@stof It would be possible to add a switch to `FormRenderer` that controls whether custom blocks are checked or not.
If this switch is disabled by default, we break BC. If this switch is enabled by default, it will be pretty useless. People will start designing away for custom blocks, and once they want to improve performance, they can't turn off the switch anymore because it would require too many changes.
---------------------------------------------------------------------------
by stof at 2012-07-17T08:08:38Z
@fabpot what do you think about it ?
---------------------------------------------------------------------------
by bschussek at 2012-07-17T08:41:43Z
Another option that just came to mind is to remove inheritance checks for anything but _widget and _row. I.e., if we render `entity_widget`, check
```
_id_widget
entity_widget
choice_widget
form_widget
```
But if we render `entity_label`, only check
```
_id_label
form_label
```
This improves PHP Templating for **170ms** and Twig for **20ms**. We gain another **150ms** for PHP Templating and **~15ms** for Twig if we also restrict custom fields (_id_widget) to the _widget and _row suffixes (it's really hard to tweak the renderer for Twig.. I think a lot of its performance bottlenecks lie in Twig itself).
Do you have any data on how often blocks other than _widget and _row are customized for specific types/IDs?
---------------------------------------------------------------------------
by stof at 2012-07-17T09:47:38Z
Well, I think most of the time other blocks are not even customized based on the type :)
---------------------------------------------------------------------------
by Tobion at 2012-07-17T14:32:39Z
From my experience rendering the form components individually is easier and more flexible than customizing by ID or type.
But there are still use cases for customizing like library-like bundles (e.g. Bootstrap).
Commits
-------
23d8735 Added NativeFileSessionHandler to classes to compile .
12d6ae7 Removed FileSessionHandler from FrameworkExtension to stop compiling a file that does not exist.
Discussion
----------
[FrameworkBundle] Removed FileSessionHandler from FrameworkExtension to stop compiling a file that does not exist.
PR #4899 removed FileSessionHandler which caused a class not found error from FrameworkBundle after the cache was created. This PR will fix it.
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
Todo: -
License of the code: MIT
---------------------------------------------------------------------------
by stof at 2012-07-13T19:12:51Z
you should add the NativeSessionHandler class in the list instead as it replaces it
---------------------------------------------------------------------------
by tystr at 2012-07-13T19:14:29Z
+1
---------------------------------------------------------------------------
by zachbadgett at 2012-07-13T19:15:55Z
Done