Commits
-------
fbed9ff Update src/Symfony/Component/HttpKernel/HttpCache/HttpCache.php
Discussion
----------
Branch 2.0 - Correct bad HttpCache behaviour when waiting for unlock.
I read the class Symfony\Component\HttpKernel\HttpCache\HttpCache and I found something which looks like a bug lines 518-520 (in lock method) :
$wait = 0;
while (is_file($lock) && $wait < 5000000) {
usleep($wait += 50000);
}
This code can wait at maximum 50000+100000+150000+.... 4950000 µs = more than 250 seconds !
I corrected like that :
$wait = 0;
while (is_file($lock) && $wait < 5000000) {
usleep(50000);
$wait += 50000
}
This code will wait 5 sec maximum.
This is more coherent if we read the following lines of this method.
---------------------------------------------------------------------------
by arnapou at 2012-03-15T20:09:13Z
Hope I succeded to do a correct PR.
One hour of manipulation in Github for 2 lines of code let me a bitter taste on the tongue...
I was closed to tell you "Do It Yourself" ... what a waste of time...
This interface is not obvious, even if we had already worked on svn/git on *nix.
Commits
-------
eee5065 [TwigBundle] Workaround a flaw in the design of the configuration (normalization)
Discussion
----------
[TwigBundle] Workaround a flaw in the design of the configuration (norma...
...lization)
see #2823
@Seldaek please comment.
---------------------------------------------------------------------------
by Seldaek at 2012-03-09T20:52:47Z
It seems fine at first glance. I don't have time to look at it in detail right now sorry.
Commits
-------
1b395f5 Revert "Throw exception when "date_widget" option is not equal to "time_widget""
Discussion
----------
Reverts commit 3c2539 to remove exception when DateTypeType has differents date and time widgets
see https://github.com/symfony/symfony/pull/1419
Commits
-------
ed218bb Fixed an "Array to string conversion" warning when using PHP 5.4. Also affects Symfony2 master.
Discussion
----------
[Config] Fixed an "Array to string conversion" warning when using PHP 5.4
This also affects Symfony2 master
Commits
-------
17c3482 fixed timezone bug in DateTimeToTimestampTransformer
Discussion
----------
[FIX]fixed timezone bug in DateTimeToTimestampTransformer
After several trials, I found out that the original code
```php
$dateTime = new \DateTime(sprintf("@%s %s", $value, $this->outputTimezone));
```
would create a DateTime object with timezone being '0000', even though $this->outputTimezone is set to my local timezone.
so I expanded the code a bit and it's working now.
PHP Test code,
```PHP
$d = new DateTime("@1234567890 Asia/Tokyo");
echo date_format($d, 'Y/m/d H:i:s')."\n";
echo $d->getTimezone()->getName()."\n";
$d = new DateTime("now Asia/Hong_Kong");
echo date_format($d, 'Y/m/d H:i:s')."\n";
echo $d->getTimezone()->getName()."\n";
```
The output is as followed:
2009/02/13 23:31:30
+00:00
2012/03/13 03:35:55
Asia/Hong_Kong
This could be a bug of PHP,
---------------------------------------------------------------------------
by stealth35 at 2012-03-13T15:54:31Z
👍
Commits
-------
50cb486 Fixed proxy generation in the DoctrineBundle when using Doctrine >= 2.2.0
Discussion
----------
[DoctrineBundle] Fixed proxy generation with Doctrine >= 2.2.0
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
The issue here was, that the name of the generated Proxy files have changed in Doctrine 2.2.0, thus the autoloader in the DoctrineBundle stoped working.
This PR fixes this issue by applying different string manipulations to the given class name depending on the currently used Doctrine version.
---------------------------------------------------------------------------
by fabpot at 2012-03-14T07:13:23Z
Can you squash your commits before I merge? Thanks.
---------------------------------------------------------------------------
by Spea at 2012-03-14T09:33:10Z
Should I open a new PR when squashed the commits? Because I don't know what happens when I force the remote repository to push my squashed commits.
---------------------------------------------------------------------------
by stloyd at 2012-03-14T09:48:05Z
First you should rebase (normally), then squash and push with `--force`, then GH will automaticaly update this PR (if you push into same branch on which bases this PR).
---------------------------------------------------------------------------
by Spea at 2012-03-14T10:04:30Z
Yeah I knew about the ```--force``` option. I just wasn't sure what happens when I do it. Thank you!
---------------------------------------------------------------------------
by Spea at 2012-03-14T10:14:23Z
Squashed commits.
Commits
-------
93cc9ef [Validator] Remove a race condition in the ClassMetaDataFactory (fix#3217)
Discussion
----------
[Validator] Remove a race condition (fix#3217)
#3581 for 2.0
Commits
-------
878c239 Fixed autoloader leakage in tests
Discussion
----------
Doctrine autoload
The autoloader for proxies is now unregistered on shutdown to avoid
having several instances registered at the same time in tests.
Commits
-------
aa53b88 Sets _format attribute only if it wasn't set previously by the user
Discussion
----------
Sets _format attribute only if it wasn't set previously by the user.
Fixes#2653
Commits
-------
705e460 provided unmerged definition for correct help generation
45bbb5b added getNativeDefinition() to allow specifying an alternate InputDefinition for help generation
Discussion
----------
[Console] Xml output fix
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
Fixes the following tickets: #2667
Todo: add specific test
As per my comment [here](https://github.com/symfony/symfony/issues/2667#issuecomment-4431944), added the ability to provide an InputDefinition that will not be changed by merging with the Application InputDefinition..
Commits
-------
f26c1ce Fixed constraint requirements for Doctrine Common
011791d [Form] Moved the Validator component to the suggest section
Discussion
----------
Composer deps
There is no hard dependency to the Validator component in the Form, as said on Twitter when @harikt tried to use it. I kept the Locale component as a requirement as it is used by the LanguageTyep, CountryType and LocaleType which will be registered when using the CoreExtension.
The constraints for Doctrine deps are fixed too: adding an upper bound everywhere as we don't know the future to guarantee the compatibility (and for instance, 2.0.9 and lower were not compatible with ORM 2.2 as we had to fix the bundle), and the bridge is compatible with Common 2.2 too, not only with 2.1.
I found 2 other places where the dependencies should be discussed:
- the Validator component marks a hard dependency to Doctrine Common for its annotation reader. There is a dependency only when using annotation so it should not be a hard requirement IMO but a suggestion. the issue is that the [ValidatorFactory](https://github.com/symfony/symfony/blob/master/src/Symfony/Component/Validator/ValidatorFactory.php) (not used by the framework itself) will add an annotation loader when relying on the default value of the arguments, which means that people that don't take care will need Common. Would it make sense to change the default so that Common is needed only when the user explicitly asks to use annotations ? Moving Common from require to suggest would make it easier for people using the Validator component standalone if they don't use annotations
- the Security component suggests the Finder and ClassLoader components. But these ones are only used by the [dev script](https://github.com/symfony/symfony/blob/master/src/Symfony/Component/Security/Acl/Resources/bin/generateSql.php) used to generate the SQL schema shipped in the component. Does it really make sense to list them as people cloning the component should probably never use this script (which alters the files in the component) ?
---------------------------------------------------------------------------
by fabpot at 2012-03-11T08:14:46Z
+1 for removing Doctrine Common as a required dependency for the Validator component.
+1 for removing ClassLoader and Finder from the Security suggestions.
Commits
-------
ad07a95 [BrowserKit] Fixed Client->back/forward/reload() not keeping all request attributes
Discussion
----------
[BrowserKit] Client->back/forward/reload() is not keeping all request attributes
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: 
Fixes the following tickets: This one
Todo: -
<hr>
Hello
While using the BrowserKit component with Behat, I noticed that some request attributes, such as files or body, disappeared when using `Symfony\Component\BrowserKit\Client->back/forward/reload()`.
The method used internally in these methods, Client->#requestFromRequest was badly passing the old request parameters to the new request. See the diff.
Commits
-------
d2f8aa3 Allow autoload to run without vendors being cloned
Discussion
----------
[Tests] Allow autoload to run without vendors being cloned
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
Commits
-------
8fa2ff0 [FrameworkBundle] fix DependencyInjection/*FrameworkExtensionTest for Windows
Discussion
----------
[FrameworkBundle] fix DependencyInjection/*FrameworkExtensionTest for Windows
Commits
-------
caa44ae Only work with the cli sapi
e2fc3cd [Process] PHP_BINARY return the current process
Discussion
----------
[Process] PHP_BINARY only return the current sapi process
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: yes
Fixes the following tickets: -
Todo: -
PHP_BINARY return the current process, so under the apache module I get (on Windows) `httpd.exe`
---------------------------------------------------------------------------
by vicb at 2012-03-02T14:48:04Z
Could it be '.../php5' ?
Is it a PHP bug ?
---------------------------------------------------------------------------
by stealth35 at 2012-03-02T15:00:57Z
@vicb
> Could it be '.../php5' ?
Yes, but it's not the standard installation, you also call you php executable 'blabla.exe'
> Is it a PHP bug ?
I'don't now, I think it's get the SAPI but If it's work under Unix so yes it's a PHP Bug
http://lxr.php.net/opengrok/xref/PHP_5_4/main/main.c#binary_location
I'll test this on my Mac
---------------------------------------------------------------------------
by vicb at 2012-03-02T15:08:25Z
@stealth35
* asking about 'php5' because it seems some systems use this name,
* would testing for a cli sapi be more robust than making assumptions about the process name ?
---------------------------------------------------------------------------
by stealth35 at 2012-03-02T15:24:24Z
@vicb yes I think it's better
Anybody can test `PHP_BINARY` with some sapi like : `cgi`, `cgi-fcgi`, `fpm-fcgi` ... ?
Commits
-------
515b581 [FrameworkBundle] Fixed test
Discussion
----------
[FrameworkBundle] Fixed test on Ubuntu 10.04 LTS
Bug fix: yes
Feature addition: no
Backwards compatibility break: no
Symfony2 tests pass: [](http://travis-ci.org/michal-pipa/symfony)
Fixes the following tickets: -
Todo: -
Test was failing on Ubuntu 10.04 LTS because Ubuntu adds it's own PHP extra version number, which was greater than 5.3.2.
The same bug exist in master branch.
Commits
-------
5a6ce20 [Session] Add Test for PDO Session Storage with SQLite in Memory DB.
Discussion
----------
[Session] PDO Session Storage tests
The PDO Session storage was untested previously.
This test is for the 2.0 API. The methods names all changed in the master branch and have to be adjusted when 2.0 is merged into master.
