This PR was submitted for the 2.3 branch but it was merged into the 2.5-dev branch instead (closes#9374).
Discussion
----------
Change of scope
Hi,
when overriding the Symfony RoleHierarchy it would be great to be able to get access to the buildRoleMap-method and map-variable for more advanced usage.
Don't hesitate to ask for further clarification if necessary - thanks in advance for merging the PR!
Kind regards,
David
Commits
-------
621aaf1 Change of scope
When overriding the Symfony RoleHierarchy it would be great to be able to get access to the buildRoleMap-method and map-variable for more advanced usage.
This PR was merged into the 2.5-dev branch.
Discussion
----------
[Validator] Added Doctrine cache
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | yes
| Tests pass? | yes
| Fixed tickets | https://github.com/symfony/symfony/issues/9887
| License | MIT
| Doc PR |
I propose to keep the `CacheInterface` and deprecate only the `ApcCache`.
It will leave the classes depending on a `CacheInterface` unchanged and will allow to adapt new cache providers in the future.
Commits
-------
3c4de45 [Validator] Added Doctrine cache
This PR was submitted for the 2.3-dev branch but it was merged into the 2.3 branch instead (closes#9896).
Discussion
----------
[Intl] Skip tests that need full lib-intl when there's no lib-intl
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
`NumberFormatter::formatCurrency` executes indirectly `Symfony\Component\Intl\ResourceBundle\Reader::read()`, which has `$bundle = new \ResourceBundle($locale, $path)` - there's Fatal Error if `\ResourceBundle` isn't found. Added availability to run unit tests if `lib-intl` isn't installed.
Commits
-------
6614b66 Skips test that need full lib-intl.
* 2.4:
fix some cs
use restore_error_handler instead of set_error_handler($previous)
fix#9321 Crawler::addHtmlContent add gbk encoding support
[Console] fixed column width when using the Table helper with some decoration in cells
[Security] Fixed problem with losing ROLE_PREVIOUS_ADMIN role.
Fix for cache-key conflict when having a \Traversable as choices
[Security] removed obsolete comment
* 2.3:
fix some cs
use restore_error_handler instead of set_error_handler($previous)
fix#9321 Crawler::addHtmlContent add gbk encoding support
[Console] fixed column width when using the Table helper with some decoration in cells
[Security] Fixed problem with losing ROLE_PREVIOUS_ADMIN role.
Fix for cache-key conflict when having a \Traversable as choices
[Security] removed obsolete comment
Conflicts:
src/Symfony/Component/Console/Helper/TableHelper.php
src/Symfony/Component/Security/Tests/Http/Firewall/ExceptionListenerTest.php
This PR was merged into the 2.3 branch.
Discussion
----------
[Console] fixed column width when using the Table helper with some decoration in cells
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #8152, #9366
| License | MIT
| Doc PR | n/a
This PR fixes the same issue as #9366 but works in all situations (all kind of styles, when the string is shorter than any other one or larger than any other ones, ...).
I'm not very satisfied with the fix and especially the `computeLengthWithoutDecoration` method, but the whole helper should be rethought to make it stateless (out of the scope of this PR).
Commits
-------
5b4d057 [Console] fixed column width when using the Table helper with some decoration in cells
This PR was merged into the 2.5-dev branch.
Discussion
----------
WebTestCase: Assume relative KERNEL_DIR is relative to phpunit.xml[.dist]?
When using the `KERNEL_DIR` setting in phpunit.xml[.dist] files for the `WebTestCase`, a relative path seems to be interpreted relative to the cwd.
That makes no difference in the probably most common case of having the phpunit config in the project's top level directory and just running `phpunit`. It makes a difference however when running it from a lower level dir and referencing the config with `-c ../../phpunit.xml`.
A conservative change would be to interpret the `KERNEL_DIR` as relative to cwd and only try it as relative to the phpunit.xml[.dist] file if the first path does not exist.
A more consistent approach would be to always have it (in case of relative paths) being relative to the config file. That might break things for folks who intentionally start phpunit from different directories to use different kernels.
The docs seem not to say anything about how it is supposed to used, the example values provided suggest using an absolute path.
Opinions?
Commits
-------
05dc0e1 Consider KERNEL_DIR setting as relative to the PhpUnit XML file if it does not point to a directory (relative to the current cwd)
This PR was submitted for the 2.2 branch but it was merged into the 2.3 branch instead (closes#9323).
Discussion
----------
[DomCrawler]fix #9321 Crawler::addHtmlContent add gbk encoding support
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #9321
| License | MIT
| Doc PR | n/a
This is solution 1 in https://github.com/symfony/symfony/issues/9321#issuecomment-26508649
Commits
-------
30af288 fix some cs
9f20b24 use restore_error_handler instead of set_error_handler($previous)
53cb6ad [DomCrawler]fix #9321 Crawler::addHtmlContent add gbk encoding support
This PR was squashed before being merged into the 2.3 branch (closes#8997).
Discussion
----------
[Security] Fixed problem with losing ROLE_PREVIOUS_ADMIN role.
<table>
<tr>
<td><b>Q</b></td>
<td><b>A</b></td>
</tr>
<tr>
<td>Bug fix?</td>
<td>yes</td>
</tr>
<tr>
<td>New feature</td>
<td>no</td>
</tr>
<tr>
<td>BC breaks?</td>
<td>no</td>
</tr>
<tr>
<td>Deprecations?</td>
<td>no</td>
</tr>
<tr>
<td>Tests pass?</td>
<td>yes</td>
</tr>
<tr>
<td>Fixed tickets</td>
<td>#3085, #8974</td>
</tr>
<tr>
<td>License</td>
<td>MIT</td>
</tr>
<tr>
<td>Doc PR</td>
<td>n/a</td>
</tr>
</table>
Problem occurs while user is impersonated. Authentication process generates new token and doeas not preserve role ```ROLE_PREVIOUS_ADMIN```. Ex. when parameter ```security.always_authenticate_before_granting``` is enabled.
Commits
-------
a7baa3b [Security] Fixed problem with losing ROLE_PREVIOUS_ADMIN role.
This PR was submitted for the 2.3-dev branch but it was merged into the 2.3 branch instead (closes#9557).
Discussion
----------
[DoctrineBridge] Fix for cache-key conflict when having a \Traversable as choices
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #9556
| License | MIT
| Doc PR | ~
Commits
-------
1119684 Fix for cache-key conflict when having a \Traversable as choices
* 2.4:
[Security] simplified some unit tests
[Security] made code easier to understand, added some missing unit tests
[DependencyInjection] fixed InlineServiceDefinitionsPass to not inline a service if it's part of the current definition (to avoid an infinite loop)
[DomCrawler] Fixed creating form objects from form nodes.
disabled php.ini changes when using HHVM in .travis.yml
[Process] fixed HHVM support
Add support for HHVM in the getting of the PHP executable
[Security] fixed error 500 instead of 403 if previous exception is provided to AccessDeniedException
* 2.3:
[Security] made code easier to understand, added some missing unit tests
[DependencyInjection] fixed InlineServiceDefinitionsPass to not inline a service if it's part of the current definition (to avoid an infinite loop)
[DomCrawler] Fixed creating form objects from form nodes.
disabled php.ini changes when using HHVM in .travis.yml
[Process] fixed HHVM support
Add support for HHVM in the getting of the PHP executable
[Security] fixed error 500 instead of 403 if previous exception is provided to AccessDeniedException
This PR was merged into the 2.3 branch.
Discussion
----------
[Security] Fix ExceptionListener to catch correctly AccessDeniedException if is not first exception
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #9544, #8467?, #9823
| License | MIT
| Doc PR |
Same as #9823 but with some refactoring of the code and with some unit tests.
When merging to 2.4, the unit tests can be simplified a bit.
Commits
-------
172fd63 [Security] made code easier to understand, added some missing unit tests
616b6c5 [Security] fixed error 500 instead of 403 if previous exception is provided to AccessDeniedException
This PR was merged into the 2.3 branch.
Discussion
----------
[Dependencyinjection] Fixed handling of inlined references in the AnalyzeServiceReferencesPass
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #8299, #9829
| License | MIT
| Doc PR | n/a
Hopefully a better fix for #9829 (ping @jakzal). Unit tests coming soon.
In some cases `InlineServiceDefinitionsPass` replaces a Reference with a service Definition. In such scenarios `AnalyzeServiceReferencesPass` was falling into an infinite loop.
Commits
-------
d650295 [DependencyInjection] fixed InlineServiceDefinitionsPass to not inline a service if it's part of the current definition (to avoid an infinite loop)
This PR was merged into the 2.3 branch.
Discussion
----------
[DomCrawler] Fixed creating form objects from named form nodes
| Q | A
| ------------- | ---
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #9619
| License | MIT
| Doc PR | -
Commits
-------
72d6322 [DomCrawler] Fixed creating form objects from form nodes.
This PR was merged into the 2.3 branch.
Discussion
----------
disabled php.ini changes when using HHVM in .travis.yml
Commits
-------
78bfd3a disabled php.ini changes when using HHVM in .travis.yml
This PR was merged into the 2.3 branch.
Discussion
----------
Add support for HHVM in the getting of the PHP executable
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a
| License | MIT
| Doc PR | n/a
Same as #9583 but with CS fixed, on 2.3, and with an added safeguard to only change the current behavior for HHVM.
Check for a PHP_BINARY environment variable before using the pre-defined PHP_BINARY constant.
HHVM has an explicit hhvm binary and a wrapper to mimic php functionality. We were running into issues with tests that ran in their own process where using the hhvm binary does not handle php code sent to it via stdin very well. We get "Nothing to do...pass file" exceptions. Unfortunately, the PHP_BINARY is always set to the explicit binary (the php wrapper is basically a symlink to the explicit binary). So, we thought about adding a check for a PHP_BINARY environment variable as the first choice when getting the PHP binary.
Commits
-------
876482f [Process] fixed HHVM support
539de17 Add support for HHVM in the getting of the PHP executable
This PR was merged into the 2.3 branch.
Discussion
----------
fixed typos
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a
| License | MIT
| Doc PR | n/a
Hope @pborreli is not going to be too jealous about all those typo fixes ;)
Commits
-------
5ff5615 fixed typos
This PR was merged into the 2.5-dev branch.
Discussion
----------
[EventDispatcher] Added TraceableEventDispatcher from HttpKernel
| Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | yes
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
Moved the reusable parts from the `TraceableEventDispatcher` from `HttpKernel` to the same class in `EventDispatcher` to allow debugging dispatchers.
I changed the `TraceableEventDispatcherInterface` to extend `EventDispatcherInterface`, so this is a BC break but I'm not sure if a lot of people use it without actually extending it.
Commits
-------
9a90e06 [EventDispatcher] Added TraceableEventDispatcher from HttpKernel