Commits
-------
3dd3d58 [EventListener] Fix an issue with sub-requests
71bf279 cleanup
acdb325 [StopWatch] Provide a cleaner API
acd1287 [Stopwatch] rename the section event to avoid collisions
eb540be [Profiler] Allow profiling the terminate event
4ccdc53 [HttpKernel] Cleanup of PdoProfilerStorage
814876f [HttpKernel] Tweak the code of the ProfilerListener
Discussion
----------
[Profiler] Allow profiling the terminate event
This PR is mainly about allowing to profile the terminate event (i.e. see it in the timeline panel)
There are some other tweaks.
---------------------------------------------------------------------------
by vicb at 2012-02-02T14:43:20Z
please don't merge for now. good question. bad answer.
---------------------------------------------------------------------------
by vicb at 2012-02-06T15:05:46Z
While first commits were focused on problem solving, the last brings a clean API with the ability to re-open an existing section in order to add events (re-setting event origins and merging them were just hacks).
Should be ready to be merged.
_Edit: Sorry, couldn't resist adding a private helper class again!_
---------------------------------------------------------------------------
by stof at 2012-02-06T18:30:09Z
@vicb you should stop adding such classes defined in the same file. Otherwise we will have to change the CS (and to stop telling we respect the PSR-0 standard)
---------------------------------------------------------------------------
by vicb at 2012-02-06T18:33:36Z
Once again PSR-0 is about autoloading which is exactly why I do not want in such cases. CS are an other matter and yes I think they should be changed to allow this (and I am going to submit a PR right now).
The only argument I could accept is whether this class should be private or not.
---------------------------------------------------------------------------
by vicb at 2012-02-06T19:57:06Z
Thanks for your valuable feedback @stof
---------------------------------------------------------------------------
by fabpot at 2012-02-11T20:53:03Z
Have you tested it on a project? Because it breaks my simple examples (where I have some sub-requests).
---------------------------------------------------------------------------
by vicb at 2012-02-12T09:47:23Z
my bad, should be ok now.
This allows people to filter easily between 404 type of responses (that are mostly for users) and real errors in their application (where they probably want to get an email notification
Quote from HTTP (bis) spec (Part 2 - 5.1.1):
The Reason Phrase exists for the
sole purpose of providing a textual description associated with the
numeric status code, out of deference to earlier Internet application
protocols that were more frequently used with interactive text
clients. A client SHOULD ignore the content of the Reason Phrase.
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 three notification methods do not return the Event instance anymore.
notify() does not return anything
notifyUntil() returns the returned value of the event that has processed the event
filter() returns the filtered value
Upgrading your listeners:
Listeners for notify() and filter() events: nothing to change
Listeners for notifyUntil() events:
Before:
$event->setReturnValue('foo');
return true;
After:
$event->setProcessed();
return 'foo';
If you notify events, the processing also need to be changed:
For filter() notifications: the filtered value is now available as
the returned value of the filter() method.
For notifyUntil() notifications:
Before:
$event = $dispatcher->notifyUntil($event);
if ($event->isProcessed()) {
$ret = $event->getReturnValue();
// do something with $ret
}
After:
$ret = $dispatcher->notifyUntil($event);
if ($event->isProcessed()) {
// do something with $ret
}
* The register() method on all listeners has been removed
* Instead, the information is now put directly in the DIC tag
For instance, a listener on core.request had this method:
public function register(EventDispatcher $dispatcher, $priority = 0)
{
$dispatcher->connect('core.response', array($this, 'filter'), $priority);
}
And this tag in the DIC configuration:
<tag name="kernel.listener" />
Now, it only has the following configuration:
<tag name="kernel.listener" event="core.response" method="filter" priority="0" />
The event and method attributes are now mandatory.
When an object has a "main" many relation with related "things" (objects,
parameters, ...), the method names are normalized:
* get()
* set()
* all()
* replace()
* remove()
* clear()
* isEmpty()
* add()
* register()
* count()
* keys()
The classes below follow this method naming convention:
* BrowserKit\CookieJar -> Cookie
* BrowserKit\History -> Request
* Console\Application -> Command
* Console\Application\Helper\HelperSet -> HelperInterface
* DependencyInjection\Container -> services
* DependencyInjection\ContainerBuilder -> services
* DependencyInjection\ParameterBag\ParameterBag -> parameters
* DependencyInjection\ParameterBag\FrozenParameterBag -> parameters
* DomCrawler\Form -> FormField
* EventDispatcher\Event -> parameters
* Form\FieldGroup -> Field
* HttpFoundation\HeaderBag -> headers
* HttpFoundation\ParameterBag -> parameters
* HttpFoundation\Session -> attributes
* HttpKernel\Profiler\Profiler -> DataCollectorInterface
* Routing\RouteCollection -> Route
* Security\Authentication\AuthenticationProviderManager -> AuthenticationProviderInterface
* Templating\Engine -> HelperInterface
* Translation\MessageCatalogue -> messages
The usage of these methods are only allowed when it is clear that there is a
main relation:
* a CookieJar has many Cookies;
* a Container has many services and many parameters (as services is the main
relation, we use the naming convention for this relation);
* a Console Input has many arguments and many options. There is no "main"
relation, and so the naming convention does not apply.
For many relations where the convention does not apply, the following methods
must be used instead (where XXX is the name of the related thing):
* get() -> getXXX()
* set() -> setXXX()
* all() -> getXXXs()
* replace() -> setXXXs()
* remove() -> removeXXX()
* clear() -> clearXXX()
* isEmpty() -> isEmptyXXX()
* add() -> addXXX()
* register() -> registerXXX()
* count() -> countXXX()
* keys()
