when esi is enabled and internal uris are generated for esi-tags, an
attribute-array consisting entirely of null-values isn't handled correctly.
The reason is that php's `http_build_query()`-method outputs an empty string
for such arrays:
http_build_query(array('foo' => '')) == 'foo='
http_build_query(array('foo' => null)) == ''
In the latter case, the generation of an URI in `HttpKernel::generateInternalUri()`
generates an URI that could not be matched by the corresponding route (ex.
`_internal/Controller/.html` opposed to `_internal/Controller/none.html` which
should be expected).
This commit adds a possible solution as well as a simple test for this issue.
Commits
-------
9bcce9f fix tests
fc4787a fix non-extensible router
Discussion
----------
Router fix
Right now, the router is hard to overwrite (you need always a compiler pass). This commit fixes this.
---------------------------------------------------------------------------
by fabpot at 2011/07/18 01:15:36 -0700
Why do you need a complier pass to override the router?
---------------------------------------------------------------------------
by schmittjoh at 2011/07/18 01:47:47 -0700
How would you suggest to overwrite it?
Basically, I want to do something like this:
```yml
services:
router:
parent: router.default
class: MyClass
calls:
- [moreDeps, []]
```
---------------------------------------------------------------------------
by Seldaek at 2011/07/18 05:07:19 -0700
Then maybe we should somehow support redefining services with the same name while keeping the old one as parent, otherwise we need this foo.default for every service out there?
---------------------------------------------------------------------------
by fabpot at 2011/07/18 06:30:34 -0700
as @Seldeak said, why do that for the router and not all services?
---------------------------------------------------------------------------
by schmittjoh at 2011/07/18 06:38:39 -0700
I have designed the SecurityBundle this way where extension is encouraged.
---------------------------------------------------------------------------
by schmittjoh at 2011/07/18 11:15:57 -0700
I should add that this is mainly a problem for services where you still want to use the semantic configuration that is provided by the bundle. For services which are not configured by the extension, this is not so much of an issue.
Anyway, if you don't want to merge it, just close the PR. I have no problem with using a compiler pass.
---------------------------------------------------------------------------
by fabpot at 2011/07/18 11:55:11 -0700
We already have such a case with translator and translator.real. I will review the existing services to see where it makes sense to implement the same strategy.
---------------------------------------------------------------------------
by Seldaek at 2011/07/18 12:20:55 -0700
I guess you'd do it anyway, but we should pick a winner between .real and .default
---------------------------------------------------------------------------
by lsmith77 at 2011/07/18 12:26:52 -0700
I would prefer ".default" as ".real" always confused me.
Commits
-------
5d46e63 [Form] Add the FormHelper configuration
a43fad4 [Form] Improve unit tests for rendering
1cb2129 [FrameworkBundle][Form] Adding a cache to FormHelper::lookupTemplate()
f39ce67 [Form][FrameworkBundle] PHP theming
Discussion
----------
[2.1] RFC [Form] Php theming
This PR implements theming support for the php engine.
It works similarly as the twig theming with themes being folders and blocks being individual files.
There are probably a few things to tune before this can get merged:
### Theme naming
The current format is "\<Bundle\>:\<Controller\>" i.e. "FrameworkBundle:Form".
Is this ok or could you imagine something better ?
### Div and Table theme folders
Currently "FrameworkBundle\\Resources\\views\\Form" and "FrameworkBundle\\Resources\\views\\FormTable"
Is this ok or anything better ?
### Form helper configuration
I am not sure if the configuration is at the best possible location:
```
framework:
templating:
form:
resources: [themeA, themeB]
```
Any better idea ?
There is a [thread on the ml](http://groups.google.com/group/symfony-devs/browse_thread/thread/9b3f131fe116b511)
Commits
-------
2c1108c [Form] Revert the ability to override anything else than the text of the label while rendering a row
da467a6 [Form] Fix the exception message when no block is found while rendering
8670995 [Form] Optimize rendering when the block to render is known
41e07c9 [Form] Optimize rendering
ee5d975 [Form] Remove a test which is no more relevant (after recent FileType refactoring)
f729c6b [Form] Add the ability to override label & widget options when rendering a row
e09ae3f [Form][FrameworkBundle] Make FormHelper::renderSection() recursively callable, introduce FormHelper::renderBlock()
e43fb98 [Form][TwigBridge] Make FormExtension::render() recursively callable to ease theming
Discussion
----------
[Form] Some refactoring of the rendering
# First two commits
## FormExtension::render() can now be called recursively.
The main use case is theming support in for collections. Let's consider that you have a collection of `CustomType`, the type hierarchy while rendering the proto would be `field < form < custom < prototype`. Before this change any theme applied to your custom type (i.e. a `custom_row` block) would not have been taken into account while rendering the prototype because of the structure of the `prototype_row` block:
```html
{% block prototype_row %}
{% spaceless %}
<script type="text/html" id="{{ proto_id }}">{{ block('field_row') }}</script>
{% endspaceless %}
{% endblock prototype_row %}
```
which skip the `custom_row` block rendering to fallback to the `field_row` block rendering.
With this PR `prototype_row` recursively calls `FormExtension::render()`
```html
{% block prototype_row %}
{% spaceless %}
<script type="text/html" id="{{ proto_id }}">{{ form_row(form) }}</script>
{% endspaceless %}
{% endblock prototype_row %}
```
this has for effect to render the block for the parent type (i.e. `custom_row`)
## FormHelper
The `FormHelper` has been updated to more closely match the `FormExtension` architecture and the templates have been modified accordingly. `echo $view['form']->renderBlock(<block name>)` is the php equivalent of `{{ block(<block name>) }}`.
The attributes are now rendered using a template rather than by the `FormHelper::attributes()` method.
Several templates have been fixed.
# Third commit
The `$varStack` property was used to forward options to the label and the widget when rendering a row. The implementation was not working as expected. The proposed way to override label and widget options is to pass these options in the `label` and `widget` keys while callinf `render_row`.
That would be:
`{{ form_row(form.field, {"attr": {<row attributes>}, "label" : {"label": <text>, "attr": {<label attr>}}, "widget" : { "attr" : {<widget attributes}} } }}`
So there is now the ability to set attributes for the row (`<div>` or `<tr>`).
This has been discussed on [the mailing list](http://groups.google.com/group/symfony-devs/browse_thread/thread/17754128ba480545). **I would like to find a compromise with @Seldaek before this gets merged**
The `$varStack` property is now only used when recursively calling `FormExtension::render()`
# Notes
I have preferred to submit several commits in order to ease review and to keep some history.
---------------------------------------------------------------------------
by stof at 2011/06/20 05:20:56 -0700
@vicb On a side note, do you think it would be possible to support form theming in PHP templates too ? Currently, the only way to customize the rendering of forms when using PHP templates is to overwrite the FrameworkBundle's templates, and this impacts all forms. This makes the PHP rendering far less powerful than the Twig one.
I don't know the Form rendering and the PHPEngine well enough to know if it is feasible for 2.1 or not.
---------------------------------------------------------------------------
by vicb at 2011/06/20 05:35:11 -0700
@stof I hope to make it possible but I need a little bit more thinking to find the best possible solution which should not look like a hack.
---------------------------------------------------------------------------
by vicb at 2011/06/21 01:13:10 -0700
This should not be merged yet, it might have some issue with the variable stack. I am working on it.
---------------------------------------------------------------------------
by vicb at 2011/06/21 01:41:11 -0700
Sorted out the issue, it was linked to some local _optimization_, the code of this PR is ok.
---------------------------------------------------------------------------
by vicb at 2011/06/21 02:01:24 -0700
I have pushed a [POC of php theming based on this PR](https://github.com/vicb/symfony/commits/form%2Fphp-theme) to my repo - it is lacking a configuration and cache layer.
I have open [a thread on the ml](http://groups.google.com/group/symfony-devs/browse_thread/thread/9b3f131fe116b511) to discuss this.
---------------------------------------------------------------------------
by vicb at 2011/06/21 23:40:21 -0700
@fabpot fixed in the last commit.
Here are the new simplified rules:
* Required cache warmers are *always* executed when the Kernel boots for the first time;
* Optional cache warmers are *only* executed from the CLI via cache:warmup
These new rules means that all the configuration settings for the cache
warmers have been removed. So, if you want the best performance, remember to
warmup the cache when going to production.
This also fixed quite a few bugs.
* gordonslondon/http-foundation/response:
[HttpFoundation] merge Response::isRedirected() with Response::isRedirect() - Response::isRedirected() has been removed
* Seldaek/events:
[EventDispatcher] Removed temporary code
[FrameworkBundle] Improved code readability
[FrameworkBundle] Clarified code and fixed regression
Update Core and Security events to latest model
[EventDispatcher] Allow registration of arbitrary callbacks
[EventDispatcher] Remove useless code
[EventDispatcher] Minor memory optimization to getListeners()
[FrameworkBundle] Small optimization, remove some function calls
This in effect removes the direct link between event name and the method name on the handler.
Any callback can be given as a handler and the event name becomes an arbitrary string. Allowing for easier namespacing (see next commit)
The consequence of this commit is that variables are accessible that have been passed to a surrounding form helper.
Example template:
{% block my_widget_label %}
<label>{{ label }}
{% endblock %}
{% block my_widget_row %}
{# It is not necessary to explicitely pass through the label variable #}
{{ form_label(form) }}
{{ form_widget(form) }}
{% endblock %}
Example usage:
{{ form_row(form.mywidget, { 'label': 'My Widget' }) }}
* vicb/service-event:
[FrameworkBundle] Optimization of the method ContainerAwareEventDispacther::dispatch()
[FrameworkBundle] Fix an issue with ContainerAwareEventDispatcher when re-entering a scope
* made the options array only for "global" options that are valid for all session storages
* changed the PDO session storage constructor signature to accept an array of options for DB configuration
* changed the storage_id to be the full service id, instead of just part of it
* removed the class parameter for session as it can be changed via the .class parameter (it was the only example in the framework)
* removed the configuration for the PDO session storage for now
* vicb/event_debug:
[FrameworkBundle] Make some TraceableEventDispacther properties private
[Event] Tweak phpDoc for consistency
[FrameworkBundle] Add error detection to the TraceableEventDispatcher
* vicb/locate_template2:
[FrameworkBundle] Enforce templates instances of TemplateReferenceInterface
[FrameworkBundle] Add unit tests for the CacheTemplateLocator class
[FrameworkBundle] Add unit tests for the TemplateLocator class
[TwigBundle] Fix the cache warmer
[TwigBundle] Tweak cache warmer configuration
[FrameworkBundle] Fix resource inheritance in the template cache warmer
The parameter has been removed and the service moved to the XML file (for consistency).
The behavior is still the same as before as any non-public service
which is not referenced anywhere will be automatically removed by a
compiler pass.
Controllers:
"BlogBundle:Post:show" is now "Blog:Post:show"
Templates:
"BlogBundle:Post:show.html.twig" is now "Blog:Post:show.html.twig"
Resources:
"@BlogBundle/Resources/config/blog.xml" is now "@Blog/Resources/config/blog.xml"
Doctrine:
"$em->find('BlogBundle:Post', $id)" is now "$em->find('Blog:Post', $id)"
It's a detail, but it hits usability. For normal bundles (those without children), we're able to actually print the namespace where we're looking for the Controller. For bundles with children, this would be a very verbose message, but we can at least print all of the bundles that we looked inside of.
The onCore* events are fired at some pre-defined points during the
handling of a request. At this is more important than the fact
that you can change things from the event.
The only missing part is ContainerAwareEventManager::addEventSubscriberService(),
because I'm not sure how to find out the class name of a service in the DIC.
Also, inline documentation of this code needs to be finished once it is accepted.
Doctrine's EventManager implementation has several advantages over the
EventDispatcher implementation of Symfony2. Therefore I suggest that we
use their implementation.
Advantages:
* Event Listeners are objects, not callbacks. These objects have handler
methods that have the same name as the event. This helps a lot when
reading the code and makes the code for adding an event listener shorter.
* You can create Event Subscribers, which are event listeners with an
additional getSubscribedEvents() method. The benefit here is that the
code that registers the subscriber doesn't need to know about its
implementation.
* All events are defined in static Events classes, so users of IDEs benefit
of code completion
* The communication between the dispatching class of an event and all
listeners is done through a subclass of EventArgs. This subclass can be
tailored to the type of event. A constructor, setters and getters can be
implemented that verify the validity of the data set into the object.
See examples below.
* Because each event type corresponds to an EventArgs implementation,
developers of event listeners can look up the available EventArgs methods
and benefit of code completion.
* EventArgs::stopPropagation() is more flexible and (IMO) clearer to use
than notifyUntil(). Also, it is a concept that is also used in other
event implementations
Before:
class EventListener
{
public function handle(EventInterface $event, $data) { ... }
}
$dispatcher->connect('core.request', array($listener, 'handle'));
$dispatcher->notify('core.request', new Event(...));
After (with listeners):
final class Events
{
const onCoreRequest = 'onCoreRequest';
}
class EventListener
{
public function onCoreRequest(RequestEventArgs $eventArgs) { ... }
}
$evm->addEventListener(Events::onCoreRequest, $listener);
$evm->dispatchEvent(Events::onCoreRequest, new RequestEventArgs(...));
After (with subscribers):
class EventSubscriber
{
public function onCoreRequest(RequestEventArgs $eventArgs) { ... }
public function getSubscribedEvents()
{
return Events::onCoreRequest;
}
}
$evm->addEventSubscriber($subscriber);
$evm->dispatchEvent(Events::onCoreRequest, new RequestEventArgs(...));
The Response is not available in the DIC anymore.
When you need to create a response, create an instance of
Symfony\Component\HttpFoundation\Response instead.
As a side effect, the Controller::createResponse() and Controller::redirect()
methods have been removed and can easily be replaced as follows:
return $this->createResponse('content', 200, array('foo' => 'bar'));
return new Response('content', 200, array('foo' => 'bar'));
return $this->redirect($url);
return Response::createRedirect($url);
How to upgrade?
For XML configuration files:
* All extensions should now use the config tag (this is just a convention as
the YAML configurations files do not use it anymore):
* The previous change means that the doctrine and security bundles now are
wrapped under a main "config" tag:
<doctrine:config>
<doctrine:orm />
<doctrine:dbal />
</doctrine:config>
<security:config>
<security:acl />
...
</security:config>
For YAML configuration files:
* The main keys have been renamed as follows:
* assetic:config -> assetic
* app:config -> framework
* webprofiler:config -> web_profiler
* doctrine_odm.mongodb -> doctrine_mongo_db
* doctrine:orm -> doctrine: { orm: ... }
* doctrine:dbal -> doctrine: { dbal: ... }
* security:config -> security
* security:acl -> security: { acl: ... }
* twig.config -> twig
* zend.config -> zend
* Remove redundant null/true equivalent array() values for array nodes
* Profiler matcher should not be deep merged; subsequent configs can simply overwrite its array
* Per lsmith's suggestion, change "isset(x) && x" to "!empty(x)"
* Templating engines node should be required, which is necessary to ensure requiresAtLeastOneElement() applies to its prototype children
In routing files, import statements allow an optional "type" option to hint the resources' type (e.g. for ambiguous file extensions). This adds the same type option to the FrameworkExtension config, which defines the main routing resource.
This reverts commit f53080860a.
Revert "[Router] config fixes"
This reverts commit 51beecc6f2.
Revert "moved duplicated files to a new Config component"
This reverts commit a8ec9b27f0.
This fixes some BC problems introduced in f9138d313b. Some top-level can now be simply enabled by providing true/null in PHP/YAML. Additionally, the Configuration\Builder allows options to be unset by providing "false" (helpful for overriding activation in a previous config file). All options supporting these behaviors can be found in the Configuration.php file (look for canBeUnset() and treatNull/TrueLike()).
Major changes:
* Removed "enabled" option for profiler config. Profiler is now enabled if its config is true, null or a map.
* Restore original config structure for validation namespaces. In PHP/YAML, namespaces are defined under annotations as an alternative to false (disabled) and true/null (enabled). For XML, annotation remains a boolean attribute for validation and a one or more optional namespace tags may appear within <app:validation />. During config normalization, namespace tags under validation will be moved to annotations to conform to the PHP/YAML structure (this occurs transparently to the user).
* Restore behavior for router/templating config sections being optional (as shown in changes to session/validation test fixtures). If either top-level section is unset in the configuration, neither feature will be enabled and the user will no longer receive exceptions due to missing a resource option (router) or engines (templating). Resource/engines will still be properly required if the respective feature is enabled.
* Remove unused router type option from XML config XSD. Type is only relevant for import statements, so this option is likely useless.
Additional small changes:
* Added isset()'s, since config options may be unset
* Wrap registerXxxConfiguration() calls in isset() checks
* Load translation.xml in configLoad(), since it's always required
* Default cache_warmer value (!kernel.debug) is determined via Configuration class
Things to be fixed:
* Configuration\Builder doesn't seem to respect isRequired() and requiresAtLeastOneElement() (or I haven't set it properly); this should replace the need for FrameworkExtension to throw exceptions for bad router/templating configs
* The config nodes for session options don't have the "pdo." prefix, as dots are not allowed in node names. To preserve BC for now, the "pdo." prefix is still allowed (and mandated by XSD) in configuration files. In the future, we may just want to do away with the "pdo." prefix.
* Translator has an "enabled" option. If there's no use case for setting "fallback" independently (when "enabled" is false), perhaps "enabled" should be removed entirely and translator should function like profiler currently does.
* Profiler matcher merging might need to be adjusted so multiple configs simply overwrite matcher instead of merging its array keys.
Restructured config format to make processing more straightforward. Important changes that might break existing configs:
* Added "enabled" option for translator (improves multi-format compat)
* Removed hash variation of validation annotations option (only boolean)
* Moved namespace option directly under validation (improves multi-format compat)
The new merge process depends on an internal array of all supported options and their default values, which is used for both validating the config schema and inferring how to merge options (as an added benefit, it helps make the extension self-documenting). Exceptions will now be thrown for merge errors resulting from unrecognized options or invalid types. Since incoming configurations are all merged atop the defaults, many isset() checks were removed. As a rule of thumb, we probably only want to ignore null values when an option would be used to set a parameter.
Also:
* Added missing attributes to symfony-1.0.xsd
* profiler: added only-exceptions attribute
* session: fix types and add pdo attributes
* Create FrameworkExtension tests with PHP/XML/YAML fixtures
* Use "%" syntax instead of calling getParameter() within FrameworkExtension
* Normalize config keys and arrays with helper methods for PHP/XML/YAML compatibility
Earlier changes:
* Remove nonexistent "DependencyInjection/Resources/" path from XmlFileLoaders
* Remove hasDefinition() checks, as register methods should only execute once
* Remove first-run logic from registerTranslatorConfiguration(), as it is only run once
* Removed apparently obsolete clearTags() calls on definitions for non-enabled features
You must now explicitly register the templating engine you want to use:
<app:templating>
<app:engine id="twig" />
</app:templating>
app.templating:
engines: ['twig']
Symfony2 comes with two such engines: 'twig', and 'php'.