This PR was merged into the master branch.
Commits
-------
cdf1d72 [FrameworkBundle] fixed requirement of the _controller palceholder for the proxy route (closes#6783)
Discussion
----------
[FrameworkBundle] fixed requirement of the _controller palceholder for the proxy route (closes#6783)
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #6783
| License | MIT
| Doc PR | n/a
---------------------------------------------------------------------------
by vicb at 2013-01-18T10:23:06Z
What about a UT ?
---------------------------------------------------------------------------
by vicb at 2013-01-18T11:28:41Z
and the syntax is wrong also !
---------------------------------------------------------------------------
by gimler at 2013-01-21T19:59:57Z
same problem the sonata admin bundle use
```
{% render 'sonata.admin.controller.admin:getShortObjectDescriptionAction' %}
```
rewrite to
```
{% render controller('sonata.admin.controller.admin:getShortObjectDescriptionAction') %}
```
throws
```
An exception has been thrown during the rendering of a template ("Parameter "_controller" for route "_proxy" must match "[^/\.]++" ("sonata.admin.controller.admin:getShortObjectDescriptionAction" given) to generate a corresponding URL.") in "SonataAdminBundle:CRUD:edit.html.twig".
```
with the requirement fix it throws
```
An exception has been thrown during the rendering of a template ("Unable to parse the controller name "sonata".") in "SonataAdminBundle:CRUD:edit.html.twig".
```
---------------------------------------------------------------------------
by fabpot at 2013-01-22T06:40:14Z
ok, I've updated the patch. There is now a static segment (`/for`) between the controller and the format, which should fix the problem.
While thinking about this, there is another option, which might be even better: removing the need for the proxy route altogether and check for a defined path like `/_proxy`. It would remove the dependency on a Url Generator in the rendering strategy, and would not make the router proxy listener any more complex.
---------------------------------------------------------------------------
by gimler at 2013-01-22T07:20:43Z
+1 for me the patch works i will open a PR for sonata doctrine orm bundle
```
{% render controller('sonata.admin.controller.admin:getShortObjectDescriptionAction', {}, {
'code': sonata_admin.field_description.associationadmin.code,
'objectId': sonata_admin.field_description.associationadmin.id(sonata_admin.value),
'uniqid': sonata_admin.field_description.associationadmin.uniqid
})
```
---------------------------------------------------------------------------
by gimler at 2013-01-22T07:22:21Z
When the proxy route is nessesary we should add a note into the upgrade guide.
+1 for less complexesy
---------------------------------------------------------------------------
by fabpot at 2013-01-22T08:02:12Z
There is one issue with removing the proxy route: when generating a proxy URL, we need a Request instance, which is not always the case. I'm going to submit another PR to "fix" that first.
---------------------------------------------------------------------------
by vicb at 2013-01-22T08:17:51Z
> It would remove the dependency
Paul leaves this body :)
---------------------------------------------------------------------------
by Tobion at 2013-01-22T08:53:52Z
I don't think removing the proxy route is good. That's the purpose of the routing system to handle generating and matching. Now if you do it manually it will probably show a bad approach to people to handle such stuff.
Also people cannot see what routes are defined explicitly and use tools like `router:debug`.
---------------------------------------------------------------------------
by fabpot at 2013-01-22T09:28:55Z
@Tobion: see #6829
---------------------------------------------------------------------------
by fabpot at 2013-01-22T09:57:57Z
I've again changed the route pattern to avoid any possible problems (even if a controller contains a `/`).
---------------------------------------------------------------------------
by Tobion at 2013-01-22T10:16:03Z
Can a controller contain `/`? It's neither a valid service nor a valid class name or?
---------------------------------------------------------------------------
by mvrhov at 2013-01-22T10:40:26Z
AFAIK yes, at least I used Namespace/SubController more then once...
* 2.0:
[DependencyInjection] fixed the creation of synthetic services in ContainerBuilder
[Security] PHPDoc in SecurityEvents
[FrameworkBundle] fixed Client::doRequest that must call its parent method (closes#6737)
[Yaml] fixed ignored text when parsing an inlined mapping or sequence (closes#6786)
[Yaml] fixed#6773
[Yaml] fixed#6770
bumped Symfony version to 2.0.23-DEV
Conflicts:
src/Symfony/Component/DependencyInjection/ContainerBuilder.php
src/Symfony/Component/HttpKernel/Kernel.php
src/Symfony/Component/Yaml/Inline.php
src/Symfony/Component/Yaml/Tests/InlineTest.php
This PR was merged into the 2.0 branch.
Commits
-------
4119caf [DependencyInjection] fixed the creation of synthetic services in ContainerBuilder
Discussion
----------
[DependencyInjection] fixed the creation of synthetic services in ContainerBuilder
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a
| License | MIT
| Doc PR | n/a
---------------------------------------------------------------------------
by stof at 2013-01-22T00:14:29Z
👍
This PR was merged into the master branch.
Commits
-------
94f6116 renamed hostname to host in the routing system (closes#6775)
001734a [Validator] fixed phpdoc
18b9e68 [HttpFoundation] renamed hostname to host in the test to be consistent
Discussion
----------
Hostname rename to host
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no (does not exist in 2.1)
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #6775
| License | MIT
| Doc PR | symfony/symfony-docs#2165
see #6775
---------------------------------------------------------------------------
by WouterJ at 2013-01-21T18:44:20Z
+1
As explained in #6775, this has been done for the following reasons:
1. It's also Request::getHost()
2. The term hostname has been obsoleted in
http://tools.ietf.org/html/rfc3986#appendix-D.2 and uses the host only
3. hostname in the RFC was defined as the registered domain name, but we
probably also want to match IP-Adresses with the pattern which is the
host = IP-literal / IPv4address / reg-name for.
This PR was squashed before being merged into the 2.0 branch (closes#6818).
Commits
-------
598ae9d [Security] PHPDoc in SecurityEvents
Discussion
----------
[Security] PHPDoc in SecurityEvents
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
I'm not sure about the description given.
This PR was merged into the master branch.
Commits
-------
aadefd3 [HttpKernel] refactored the HTTP content renderer to make it easier to extend
Discussion
----------
[HttpKernel] refactored the HTTP content renderer to make it easier to extend
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | kinda
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a
| License | MIT
| Doc PR | n/a
This makes the StreamedResponse logic reusable for other strategies and it also makes the RenderingStrategy interface less fuzzy about its contract.
That also makes features like #4470 easier to implement from the outside.
---------------------------------------------------------------------------
by stof at 2013-01-20T11:01:29Z
👍
This PR was merged into the master branch.
Commits
-------
32322a1 [TwigBridge] fixed format_file to include the line number even if the link text is passed
Discussion
----------
[TwigBridge] fixed format_file to include the line number even if the link text is passed
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a
| License | MIT
| Doc PR | n/a
That fixes the logs in the profiler where lines were not displayed for deprecated calls.
---------------------------------------------------------------------------
by stof at 2013-01-20T10:58:58Z
shouldn't this be merged in older branches as it is a bugfix ?
---------------------------------------------------------------------------
by fabpot at 2013-01-20T15:14:50Z
It is a bug fix but people might rely on the current behavior. But then, nobody ever reported it. So, I prefer to keep it on 2.2 only.
This PR was merged into the master branch.
Commits
-------
1ab48db tweaked the deprecated error handler to actually display the deprecation message
Discussion
----------
tweaked the deprecated error handler to actually display the deprecation message
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a
| License | MIT
| Doc PR | n/a
I've also removed uninterested information (like the trigger_error stack
frame).
This PR was merged into the master branch.
Commits
-------
fcabadf Fix JsonDecode to work on PHP 5.3, update the CHANGELOG.md
b6bdb45 Completly refactor the Serializer Options Pull Request to push context information directly and avoid state and dependencies between SerializerInterface and encoders/normalizers.
ef652e2 Added context to JsonEncoder
eacb7e2 Rename $options to $context, as it makes the intent much more clear.
8854b85 Fix CS issues, removed global options
9c54a4b [Serializer] Allow options to be passed to SerialiizerInterface#serialize and #unserialize. Thsee options are available to all encoders/decoders/normalizers that implement SerializerAwareInterface.
Discussion
----------
[2.2] [Serializer] Configurable Serializer
Bug fix: no
Feature addition: yes
Backwards compatibility break: yes
Symfony2 tests pass: yes
Fixes the following tickets: #4907, #4938
License of the code: MIT
Todo:
This is an extension of GH-6574 that removes the context state in favor of passing this information around.
---------------------------------------------------------------------------
by beberlei at 2013-01-18T13:12:39Z
@fabpot @lsmith I think this is how it should work from an OOP/OOD perpesctive, avoiding the context state. This makes for a much cleaner code and dependency graph.
---------------------------------------------------------------------------
by lsmith77 at 2013-01-18T14:14:37Z
makes sense. anything fancier would lose this components simplicity which IMHO is the main benefit versus JMS serializer.
---------------------------------------------------------------------------
by fabpot at 2013-01-18T14:26:25Z
Looks very good. 👍
---------------------------------------------------------------------------
by beberlei at 2013-01-18T14:37:32Z
I need to fix the failures with the JsonEncoder and then this is good to merge
---------------------------------------------------------------------------
by stof at 2013-01-18T14:40:21Z
you also need to update the CHANGELOG of the component
---------------------------------------------------------------------------
by beberlei at 2013-01-18T23:17:57Z
Fixed, only the Redis Profiler problem still failing the Travis builds. Also I updated the CHANGELOG.md.
@fabpot Good to merge from my POV
---------------------------------------------------------------------------
by stof at 2013-01-18T23:27:59Z
@beberlei see #6804 for the Redis profiler issue
This PR was submitted for the master branch but it was merged into the 2.1 branch instead (closes#6803).
Commits
-------
be05741 Added an error message in the DebugClassLoader when using / instead of \.
Discussion
----------
Added an error message in the DebugClassLoader when using / instead of \.
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR | n/a
Using ``/`` instead of ``\`` when writing class names in a configuration seems to happen regularly. See aba8f1e180 (commitcomment-2461266) for the latest case I saw. I guess it may come from the fact that SensioGeneratorBundle allows using both for convenience in the CLI to get rid of escaping issues.
This will give people a better error message when they use a ``/`` instead of having them figure that the class name in the message using ``/`` and not ``\`` and this is the reason of the issue.
This PR was merged into the master branch.
Commits
-------
d5e73e5 [SecurityBundle] changed includes to use the new Twig notation
Discussion
----------
[SecurityBundle] changed includes to use the new Twig notation
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | no - couple of unrelated failures
| Fixed tickets | N/A
| License | MIT
| Doc PR | N/A
This PR was merged into the master branch.
Commits
-------
3d762dd [HttpKernel] Fixed the Redis profiler storage return value
44fe249 Fixed some tests on Windows
Discussion
----------
Fix tests
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR | n/a
This fixes the testsuite in master.
For the Redis storage, the test failure seems to come from a bad resolution when merging 2.1 into master.
---------------------------------------------------------------------------
by stof at 2013-01-18T20:58:59Z
I still have some failing tests in master locally:
- some tests with the incremental output tests in the Process component
- a timing test for the Stopwatch component (but passing when running only the Stopwatch tests)
- some mock expectations in the Form component.
However, all these tests are passing on Travis so it looks weird.
This PR was merged into the master branch.
Commits
-------
94cb13d [FrameworkBundle] fixed circular reference for service templating (closes#6730, closes#6762)
Discussion
----------
[FrameworkBundle] fixed circular reference for service templating (closes#6730, closes#6762)
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #6730, #6762
| License | MIT
| Doc PR | n/a
---------------------------------------------------------------------------
by tristanbes at 2013-01-18T14:55:39Z
@fabpot Do you have an idea when you'll merge it please ?
This PR was merged into the 2.0 branch.
Commits
-------
53ccc2c [Yaml] fixed ignored text when parsing an inlined mapping or sequence (closes#6786)
Discussion
----------
[Yaml] fixed ignored text when parsing an inlined mapping or sequence (closes#6786)
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #6786
| License | MIT
| Doc PR | n/a
This PR was merged into the 2.0 branch.
Commits
-------
ce38069 [FrameworkBundle] fixed Client::doRequest that must call its parent method (closes#6737)
Discussion
----------
[FrameworkBundle] fixed Client::doRequest that must call its parent method (closes#6737)
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #6737
| License | MIT
| Doc PR | n/A
This PR was squashed before being merged into the master branch (closes#6789).
Commits
-------
917f473 [Console] Removing unnecessary sprintf in Application->getHelp
Discussion
----------
[Console] Removing unnecessary sprintf in Application->getHelp
Minor change of the `Symfony\Component\Console\Application->getHelp()` method.
I have spotted:
1. an unnecessary `sprintf` call (no `args` arguments);
2. two ways of adding a new line in the help (an empty string as new entry of the array and a `\n` at the end of the string).
It seems to be there since the begining and it looks like a forgoten change to me, so I fixed them by removing the `sprintf` call and using a new array entry (empty string) instead of the `\n`.
| Q | A
| ------------- | ---
| License | MIT
This PR was merged into the 2.0 branch.
Commits
-------
fea20b7 [Yaml] fixed#6770
Discussion
----------
[Yaml] fixed parsing of negative integers (2.0 branch)
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #6770
| License | MIT
| Doc PR | n/a
Note that an unrelated test fixture for large integers had to be changed to work on systems with 64-bit integer support because of the change from `assertEquals()` to `assertSame()`. Please see the diff for clarification.
This PR was merged into the 2.0 branch.
Commits
-------
ab0385c [Yaml] fixed#6773
Discussion
----------
[Yaml] fixed parsing of blocks with 'keep' chomping indicators (2.0 branch)
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #6773
| License | MIT
| Doc PR | n/a
Also added a full set of tests for all combinations of literal/folded blocks and strip/clip/keep chomping indicators in both mid-file and end-of file positions.
* 2.1:
[Yaml] fixed unneeded BC break
[DependencyInjection] fixed a bug in the YAML dumper where references where not converted to the @ notation
[Yaml] fixed typo
updated VERSION for 2.1.7
updated CHANGELOG for 2.1.7
updated VERSION for 2.0.22
update CONTRIBUTORS for 2.0.22
updated CHANGELOG for 2.0.22
Conflicts:
src/Symfony/Component/HttpKernel/Kernel.php