Commits
-------
411a0cc [Validator] Added GroupSequenceProvider to changelog
815c769 [Validator] Renamed getValidationGroups to getGroupSequence
d84a2e4 [Validator] Updated test expectations
9f2310b [Validator] Fixed typos, renamed hasGroupSequenceProvider
e0d2828 [Validator] GroupSequenceProvider tests improved, configuration changed
c3b04a3 [Validator] Changed GroupSequenceProvider implementation
6c4455f [Validator] Added GroupSequenceProvider
Discussion
----------
[Validator] Added GroupSequenceProvider
Bug fix: no
Feature addition: yes
Backwards compatibility break: no
Symfony2 tests pass: 
As discussed in #3114 I implemented the "GroupSequenceProvider" pattern for the validator component. It allows the user to select certain validation groups based on the current state of an object. Here is an example:
/**
* @Assert\GroupSequenceProvider("UserGroupSequnceProvider")
*/
class User
{
/**
* @Assert\NotBlank(groups={"Premium"})
*/
public function getAddress();
public function hasPremiumSubscription();
}
class UserGroupSequenceProvider implements GroupSequenceProviderInterface
{
public function getValidationGroups($user)
{
if ($user->hasPremiumSubscription()) {
return array('User', 'Premium');
} else {
return array('User');
}
}
}
With this patch there are two mechanisms to define the group sequence now. Either you can use @GroupSequence to define a static order of validation groups or you can use @GroupSequenceProvider to create dynamic validation group arrays.
The ClassMetadata therefore has methods now which implement quite similar things. The question is whether it would make sense to interpret the static group sequence as a special case and create something like a DefaultGroupSequenceProvider or StaticGroupSequenceProvider which is assigned by default. This would cause a BC break inside the validator component.
---------------------------------------------------------------------------
by bschussek at 2012-01-28T13:39:54Z
I like the implementation, but I think we should differ a little bit from Java here.
1. `GroupSequenceProviderInterface` should be implemented by the domain classes themselves (`User`), not by a separate class.
2. As such, the parameter `$object` from `getValidationGroups($object)` can be removed
3. `ClassMetadata::setGroupSequenceProvider()` should accept a boolean to activate/deactivate this functionality. Also the check for the interface (does the underlying class implement it?) should be done here
Apart from that, special cases need to be treated:
* A definition of a group sequence and a group sequence provider in the same `ClassMetadata` should not be allowed. Either of them must not be set.
* Metadata loaders must take care of settings made by parent classes. If `Animal` is extended by `Dog`, `Animal` defines a group sequence (or group sequence provider) and `Dog` a group sequence provider (or group sequence), only the setting of `Dog` should apply
---------------------------------------------------------------------------
by blogsh at 2012-01-28T21:25:37Z
Changes of the latest commit:
- GroupSequenceProviderInterface has to be implemented by the domain class
- The annotation/configuration options let the user define whether the provider is activated or not (is this neccessary at all?)
- An error is thrown if the user wants to use static group sequences and the provider simultaneously
At the moment neither the static group sequence nor the provider is inherited from parent classes or interfaces. I don't know if it would make sense to enable this feature. There could be problems if a user wants to define a static group sequence in the parent class and a sequence provider in the child class.
---------------------------------------------------------------------------
by bschussek at 2012-01-30T13:07:04Z
> There could be problems if a user wants to define a static group sequence in the parent class and a sequence provider in the child class.
In this case, the setting in the child class should override the setting of the parent class.
But we can leave this open for now. As it seems, [this issue is unresolved in Hibernate as well](https://hibernate.onjira.com/browse/HV-467).
---------------------------------------------------------------------------
by blogsh at 2012-01-30T22:54:41Z
Okay, finally I managed to upload the latest commit. If you got a bunch of notifications or so I'm sorry, but I had to revert some accidental changes in the commit :(
I've rewritten the tests and have removed the "active" setting in the XML configuration.
---------------------------------------------------------------------------
by blogsh at 2012-02-02T15:24:01Z
Okay, typos are fixed now and `hasGroupSequenceProvider` has been renamed to `isGroupSequenceProvider`. I also had to adjust some tests after the rebase with master.
---------------------------------------------------------------------------
by bschussek at 2012-02-03T09:25:19Z
Looks good.
@fabpot 👍
---------------------------------------------------------------------------
by fabpot at 2012-02-03T09:46:52Z
Can you add a note in the CHANGELOG before I merge? Thanks.
---------------------------------------------------------------------------
by blogsh at 2012-02-09T12:31:27Z
@fabpot done
A new ExecutionContext is now created everytime that GraphWalker::walkConstraint() is
launched. Because of this, a validator B launched from within a validator A can't break
A anymore by changing the context.
Because we have a new ExecutionContext for every constraint validation, there is no point
in modifying its state anymore. Because of this it is now immutable.
Commits
-------
92f820a Renamed registerConstraints to loadDynamicValidatorMetadata
dd12ff8 CS fix, getConstraints renamed
09c1911 [Validator] Improved dynamic constraints
54cb6e4 [Validator] Added dynamic constraints
Discussion
----------
[Validator] Dynamic constraints
Bug fix: no
Feature addition: yes
Backwards compatibility break: no
Symfony2 tests pass: yes
By now the Validator component is based on a per-class configuration of
constraints, but in some cases it might be neccessary to add new constraints
dynamically at runtime.
This pull request adds a "ConstraintProviderInterface" to the Validator component. If an object is validated that implements this interface the method "getConstraints" is used to add dynamic constraints:
class User implements ConstraintProviderInterface
{
protected $isPremium;
protected $paymentInformation;
public function getConstraints(ClassMetadata $metadata)
{
if ($this->isPremium) {
$metadata->addPropertyConstraint('paymentInformation', new NotBlank());
}
}
}
---------------------------------------------------------------------------
by alexandresalome at 2012-01-15T11:20:04Z
Related to #1151
---------------------------------------------------------------------------
by canni at 2012-01-16T09:22:28Z
👍
---------------------------------------------------------------------------
by bschussek at 2012-01-16T12:32:44Z
I think this is a good addition. I think we still have a naming problem though. When constraints are loaded using a static method, the default name for the loader method is `loadValidatorMetadata`. Since the method for dynamic constraint loading is basically the same, I think the two names should be related.
Solution (1): Rename the method in your interface to `loadDynamicValidatorMetadata`. Ugly and long.
class MyClass implements ConstraintProviderInterface
{
public static loadValidatorMetadata(ClassMetadata $metadata) ...
public loadDynamicValidatorMetadata(ClassMetadata $metadata) ...
}
Solution (2): Rename the default method name in `StaticMethodLoader` to `registerConstraints` and adjust the docs. Breaks BC.
class MyClass implements ConstraintProviderInterface
{
public static registerConstraints(ClassMetadata $metadata) ...
public registerDynamicConstraints(ClassMetadata $metadata) ...
}
@fabpot: Are we allowed to break BC here? If not, we should probably stick to (1).
---------------------------------------------------------------------------
by fabpot at 2012-01-16T12:36:14Z
I would prefer to not break BC if possible.
---------------------------------------------------------------------------
by blogsh at 2012-01-16T15:25:46Z
So "loadDynamicValidatorMetadata" would be the best solution?
---------------------------------------------------------------------------
by althaus at 2012-01-17T13:39:19Z
>So "loadDynamicValidatorMetadata" would be the best solution?
Sounds fine for me based on @bschussek's comment.
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
Fixes the following tickets: #2779
Todo: -
Because PHP function `array_key_exists` is buggy, it works great with native
PHP `ArrayObject` instances, but hand written implementations of `ArrayAccess`
and `Traversable` objects will fail to work with `CollectionValidator`
Commits
-------
1e370d7 typo fix
93d8d44 added some more infos about Config
27efd59 added READMEs for the bridges
34fc866 cosmetic tweaks
d6af3f1 fixed README for Console
6a72b8c added basic README files for all components
Discussion
----------
added basic README files for all components and bridges
heavily based on http://fabien.potencier.org/article/49/what-is-symfony2 and the official Symfony2 documentation
---------------------------------------------------------------------------
by jmikola at 2011/11/03 13:36:07 -0700
Great work. For syntax highlighting on the PHP snippets, you could add "php" after the three backticks.
---------------------------------------------------------------------------
by lsmith77 at 2011/11/03 13:41:29 -0700
done
---------------------------------------------------------------------------
by stealth35 at 2011/11/03 13:49:31 -0700
Nice job, but you also need to add `<?php`
ex :
``` php
<?php
use Symfony\Component\DomCrawler\Crawler;
$crawler = new Crawler();
$crawler->addContent('<html><body><p>Hello World!</p></body></html>');
print $crawler->filter('body > p')->text();
```
---------------------------------------------------------------------------
by lsmith77 at 2011/11/03 13:56:57 -0700
done
---------------------------------------------------------------------------
by ericclemmons at 2011/11/03 19:57:57 -0700
@lsmith77 Well done! This makes consumption of individual components that much easier, *especially* now that `composer.json` files have been added.
---------------------------------------------------------------------------
by lsmith77 at 2011/11/04 01:18:23 -0700
ok .. fixed the issues you mentioned @fabpot
---------------------------------------------------------------------------
by lsmith77 at 2011/11/11 15:00:27 -0800
@fabpot anything else left? seems like an easy merge .. and imho there is considerable benefit for our efforts to spread the word about the components with this PR merged.
---------------------------------------------------------------------------
by drak at 2011/11/11 18:54:13 -0800
You know, it might be a nice idea to put a link to the documentation for each component if there is some at symfony.com
---------------------------------------------------------------------------
by lsmith77 at 2011/11/12 00:59:14 -0800
i did that in some. but i might have missed a few places.
On 12.11.2011, at 03:54, Drak <reply@reply.github.com> wrote:
> You know, it might be a nice idea to put a link to the documentation for each component if there is some at symfony.com
>
> ---
> Reply to this email directly or view it on GitHub:
> https://github.com/symfony/symfony/pull/2561#issuecomment-2715762
---------------------------------------------------------------------------
by breerly at 2011/11/21 10:28:36 -0800
Pretty excited with this.
---------------------------------------------------------------------------
by dbu at 2011/11/24 00:02:50 -0800
is there anything we can help with to make this ready to be merged?
---------------------------------------------------------------------------
by lsmith77 at 2011/12/18 02:39:23 -0800
@fabpot: seriously .. if you are not going to deliver something "better" and don't provide a reason what is wrong with this .. then its beyond frustrating. i obviously do not claim that these README's are perfect (and certainly still no replacement for proper documentation), but I do claim that in their current form they are a radical step forward to potential users of the Symfony2 components.
Commits
-------
cd24fb8 change explode's limit parameter based on known variable content
b3cc270 minor optimalisations for explode
Discussion
----------
[FrameworkBundle][CssSelector][HttpFoundation][HttpKernel] [Security][Validator] Minor optimizations for "explode" function
Bug fix: no
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
Fixes the following tickets: -
Todo: -
I added limit parameter in some places, where it may be usefull. I did not check the context of what values may have been exploded. So to not break anything, I added +1 to limit parameter.
If you find out that in some places limit (or limit+1) is not important or meaningless, write a comment please and I will fix it.
---------------------------------------------------------------------------
by fabpot at 2011/12/07 06:56:49 -0800
Adding +1 just to be sure to not break anything is clearly something we won't do. What is the benefit of doing that anyway?
---------------------------------------------------------------------------
by pulzarraider at 2011/12/07 13:50:24 -0800
The main idea of making this PR was to notify about some places that may run faster with just adding one parameter to explode function.
If in code is someting like: ```list($a, $b) = explode(':', $s);```
Function ```explode``` will create n-items (depends on ```$s```), but we need in code only the first two items. There is no reason to let ```explode``` create more items in memory that are NEVER used in our code. The limit parameter is there for these situations, so let's use it.
I know that it is microoptimization and may look unimportant, but we are writing a framework - so people expect that code will be as fast as possible without this kind of mistakes.
As I've noticed above, I know that +1 is not ideal solution, but the fastest without debugging the code. I expect that someone (with good knowledge of that code) will look at it and write in comments if variable may contain 1 comma (dot or someting on what is doing the explode) or maybe 2 in some situations or more.
Anyway, +1 will not break anything, because same items are created as it is now, but no unnecessary item is created.
---------------------------------------------------------------------------
by fabpot at 2011/12/07 23:14:59 -0800
I'm +1 for adding the number to avoid problems but I'm -1 on the optimization side of things as it won't optimize anything.
---------------------------------------------------------------------------
by helmer at 2011/12/08 12:46:49 -0800
*.. The main idea of making this PR was to notify about some places that **may** run faster ..*
I am also unsure the optimization is really an optimization, care to benchmark (with meaningful inputs)? As for the limit+1 thing, why would you want to +1 it? The number of ``list`` arguments should always reflect the ``limit`` parameter, no?
---------------------------------------------------------------------------
by pulzarraider at 2011/12/08 23:11:34 -0800
@helmer please try this simple benchmark:
```
<?php
header('Content-Type: text/plain; charset=UTF-8');
define('COUNT', 10000);
$source_string = 'aaaaaaaaaaaaaaaaaaaa:bbbbbbbbbbbbbbbbbbbbb:cccccccccccccccccccccccc:dddddddddddddddddddddd:eeeeeeeeeeeeeeeeeeeeeeeee:fffffffffffffffffffffffffff';
$start = microtime(true);
for ($i = 0; $i < COUNT; $i++) {
list($a, $b) = explode(':', $source_string);
}
$end = microtime(true)-$start;
echo 'without limit: '.$end."\n";
$start = microtime(true);
for ($i = 0; $i < COUNT; $i++) {
list($a, $b) = explode(':', $source_string, 2);
}
$end = microtime(true)-$start;
echo 'with limit: '.$end."\n";
```
My results are:
```
without limit: 0.057228803634644
with limit: 0.028676986694336
```
That is 50% difference (with APC enabled). Of course the result depends on the length of source string and if it's too short, the difference may be none or very very small. That's why I said, that it **may** run faster and is just a micro optimization.
---------------------------------------------------------------------------
by pulzarraider at 2011/12/08 23:18:12 -0800
@helmer And why +1? It depends on a code:
```
$source_string = 'aaaaaaaaaaaaaaaaaaaa:bbbbbbbbbbbbbbbbbbbbb:cccccccccccccccccccccccc';
list($a, $b) = explode(':', $source_string, 2);
var_dump($a, $b);
```
and
```
$source_string = 'aaaaaaaaaaaaaaaaaaaa:bbbbbbbbbbbbbbbbbbbbb:cccccccccccccccccccccccc';
list($a, $b) = explode(':', $source_string, 3);
var_dump($a, $b);
```
gives different results. That's why the content of the variable must be known.
---------------------------------------------------------------------------
by helmer at 2011/12/09 00:08:28 -0800
@pulzarraider Thanks for the benchmark, seems like a gain enough. Although, we are more likely having a scenario of:
``explode(':', 'a🅱️c')`` vs ``explode(':', 'a🅱️c', 3)`` with a ``COUNT`` of 10, where the difference is not even in microseconds anymore :)
The limit addition alters the behaviour though, ie suddenly you can define a controller [logical name](http://symfony.com/doc/current/book/routing.html#controller-string-syntax) as ´´AcmeBlogBundle:Blog:show:something``, and things go downhill from there on.
All that aside, I'm +1 for setting the limit to the exact number of ``list`` parameters, but certainly not number+1, this is just too wtfy (as you said, this was a safety thing, but I reckon for this PR to be merged it needs to be +0).
---------------------------------------------------------------------------
by drak at 2011/12/09 08:28:58 -0800
Overall `list()` is ugly as it's not very explicit. Even though it would mean extra lines, it's better to `explode()` then explicitly assign variables:
```
$parts = explode(':', $foo);
$name = $parts[0];
$tel = $parts[1];
```
`list()` is one of those bad relics from the PHP past...
---------------------------------------------------------------------------
by fabpot at 2011/12/11 10:07:47 -0800
@drak: why is `list` not explicit? It is in fact as explicit as the more verbose syntax you propose.
---------------------------------------------------------------------------
by pulzarraider at 2011/12/11 13:08:50 -0800
@drak: I agree with @fabpot. In speech of benchmarks ```list``` is faster then using a helper variable.
@fabpot, @helmer I've changed explode's limit to be correct (without +1) and removed some changes from this PR, where I can't find out what the content of variable may be. Unit tests pass, so I think it's ready for merge.
Commits
-------
731b28b [composer] add missing deps for FrameworkBundle
9c8f100 [composer] change ext/intl to the new ext-intl syntax
d535afe [composer] fix monolog-bridge composer.json, add more inter-component deps
9ade639 [composer] add composer.json
Discussion
----------
Composer
This PR adds a composer.json file for [composer](https://github.com/composer/composer) ([more info](packagist.org/about-composer)).
For discussion you can also go into #composer-dev on freenode and argue with naderman, seldaek and everzet.
---------------------------------------------------------------------------
by naderman at 2011/09/26 15:51:51 -0700
You haven't entered any keywords, they might come in handy when searching for packages on packagist.
But really this is just a +1 ;-)
---------------------------------------------------------------------------
by stof at 2011/09/26 16:12:21 -0700
See my comments on your previous (non-rebased) commit: f1c0242b5a
---------------------------------------------------------------------------
by igorw at 2011/09/27 00:04:36 -0700
Following dependencies do not have a composer.json yet: Twig, Doctrine (orm, dbal, common), swiftmailer.
Also missing from the standard edition: assetic, twig-extensions, jsm-metadata, SensioFrameworkExtraBundle, JMSSecurityExtraBundle, SensioDistributionBundle, SensioGeneratorBundle, AsseticBundle.
The point is, those can be added later on. Having the components composerized is already a leap forward. Also, doctrine depends on some symfony components, we've got to start somewhere.
---------------------------------------------------------------------------
by Seldaek at 2011/09/27 00:36:41 -0700
Also, just for information, the plan is to have `symfony/framework-bundle` be the "framework", with all dependencies to doctrine etc, though we should really only have strict requirements in there, and then in symfony-standard we ship a composer.json that requires the framework-bundle, doctrine-orm and things like that that are not essential to core. Otherwise people don't have a choice about what they use anymore.
Just a comment btw, the json is invalid, all / should be escaped. However json_decode is nice enough to parse those without complaining, browsers do too, even Crockford's json2.js does, so I'm not sure if we should privilege readability over strictness, since it seems nobody really cares about this escaping.
---------------------------------------------------------------------------
by igorw at 2011/09/27 00:41:39 -0700
So, I've implemented all of @stof's suggestions, except (for reasons stated above):
* doctrine to DoctrineBundle
* swiftmailer to SwiftmailerBundle
* twig to TwigBundle
* doctrine-common to Validator
* FrameworkBundle (what exactly does it depend on?)
---------------------------------------------------------------------------
by stof at 2011/09/27 00:52:31 -0700
@igorw at least HttpKernel, Routing, Templating, EventDispatcher, Doctrine Common (annotations cannot be disabled), Translator, Form (optional), Validator (optional), Console (optional). See the service definitions to see the others
@Seldaek FrameworkBundle does not depend on Doctrine, except for Common
---------------------------------------------------------------------------
by beberlei at 2011/09/27 03:15:34 -0700
What does the symfony/ or ext/ prefix control in composer?
---------------------------------------------------------------------------
by Seldaek at 2011/09/27 03:33:52 -0700
symfony/ is just the (mandatory) vendor namespace. Also ext/ has been renamed to ext- now, so it's not in any vendor, and should avoid potential issues.
---------------------------------------------------------------------------
by beberlei at 2011/09/27 05:07:03 -0700
@Seldaek Mandatory? So every package name is "vendor/package"? I like that because previously i thought package names are not namespaced, and thus clashes could occur between different communities easily.
---------------------------------------------------------------------------
by Seldaek at 2011/09/27 05:16:20 -0700
@beberlei: Mandatory. As of yesterday http://packagist.org/ will tell you you have an invalid package name if there's no slash in it. See 1306d1ca82 (diff-3)
* 2.0:
[Validator] added support for grapheme_strlen when mbstring is not installed but intl is installed
removed separator of choice widget when the separator is null
* EvanK-patch-1:
Per the [documentation][1], the `NotBlank` constraint should be using the `empty` language construct, otherwise it will not trigger on, for example, a boolean false from an unchecked checkbox field.
Commits
-------
275da0d [Validator] changed 'self' to 'static' for child class to override pattern constant
Discussion
----------
[Validator] change 'self::' to 'static::' for PATTERN constant overridable in child classes
In TimeValidator and UrlValidator, PATTERN constant is not used with late static bind(static::) while DateValidator supports it.
