This PR was merged into the 2.8 branch.
Discussion
----------
[SecurityBundle] Fix Twig master dep
| Q | A
| ------------- | ---
| Branch? | 2.8
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | yes
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Since Twig 2 is not tagged yet, it's free to do not-BC changes, which it did recently. Our unmaintained `3.0.*` bridge+bundle don't work with it anymore. Let's skip them.
Commits
-------
9a2672e [SecurityBundle] Fix Twig master dep
This PR was submitted for the master branch but it was merged into the 2.7 branch instead (closes#20077).
Discussion
----------
[Process] silent file operation to avoid open basedir issues
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #19913
| License | MIT
| Doc PR |
Using file operations on a symlink that points to a file outside the directories being allowed by the open_basedir setting leads to PHP warnings being raised.
Commits
-------
30c9b1e silent file operation to avoid open basedir issues
Using file operations on a symlink that points to a file outside the
directories being allowed by the open_basedir setting leads to PHP
warnings being raised.
This PR was merged into the 2.7 branch.
Discussion
----------
fixed Twig support for 1.26 and 2.0
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a
| License | MIT
| Doc PR | n/a
Removes the deprecation notices for Twig 1.26.
Commits
-------
24e9cf2 fixed Twig support for 1.26 and 2.0
This PR was merged into the 2.8 branch.
Discussion
----------
Fix indexBy type extraction
| Q | A
| ------------- | ---
| Branch? | 2.8+
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | N/A
| License | MIT
| Doc PR | N/A
Bug found and patched by @ksom
Since 3008228b48, the Doctrine Bridge's PropertyInfo Extractor tries to extract the type of the key in an indexed association. However, as you can see in 3008228b48 (diff-7a8fb8072d57f95ea6e37898b05895bcR91), the extractor was using the metadata of the class containing the association instead of the target entity's metadata to retrieve the type of the index.
The tests were green because in 3008228b48 (diff-c7e914ed89ceffd939733efe08c039c2R44), the property used to `indexBy` was present in the classes on both sides of the relation with the same type.
Once the test is fixed (by renaming the property in the targetEntity), the test provided at 3008228b48 (diff-1b2e044d1df011c00caf802a07895bdbR88) gives the error
1) Symfony\Bridge\Doctrine\PropertyInfo\Tests\DoctrineExtractorTest::testExtract with data set #9 ('indexedBar', array(Symfony\Component\PropertyInfo\Type Object (...)))
InvalidArgumentException: "" is not a valid PHP type.
The fix is to fetch the metadata of the target entity and check there for the property type.
Commits
-------
138c6e3 Fix indexBy type extraction
This PR was merged into the 2.7 branch.
Discussion
----------
[Console] Format JSON fixtures
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | comma-separated list of tickets fixed by the PR, if any
| License | MIT
| Doc PR | reference to the documentation PR, if any
Should make merging upstream easier due better diffs.
```php
foreach (glob('src/Symfony/Component/Console/Tests/Fixtures/*.json') as $file) {
file_put_contents($file, str_replace(PHP_EOL, "\n", json_encode(json_decode(trim(file_get_contents($file))), JSON_PRETTY_PRINT))."\n");
}
```
Should be re-applied on 2.8, 3.1 and master.
Commits
-------
2ca7823 format json fixtures
This PR was merged into the 2.7 branch.
Discussion
----------
[Finder] Trim trailing directory slash in ExcludeDirectoryFilterIterator
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #19599
| License | MIT
| Doc PR | reference to the documentation PR, if any
In this context `path` equals `path/`
Commits
-------
e0e5f0c apply rtrim
* 2.7:
Fixed expectedException annotations
Security and validators messages translation to Latvian
[Form] FormView->isRendered() remove dead code and simplify the flow
This PR was merged into the 2.8 branch.
Discussion
----------
@expectedException expects FQCN
| Q | A
| ------------- | ---
| Branch? | 2.8
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | comma-separated list of tickets fixed by the PR, if any
| License | MIT
| Doc PR | reference to the documentation PR, if any
See #20028
Almost there :)
Commits
-------
526b852 expectedException expects FQCN
This PR was squashed before being merged into the 2.7 branch (closes#20028).
Discussion
----------
Fixed expectedException annotations
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | comma-separated list of tickets fixed by the PR, if any
| License | MIT
| Doc PR | reference to the documentation PR, if any
See #19237... as this is not going to be fixed in phpunit nor php-cs-fixer and i believe by now these are the last 2 false positives.
Found with `@expectedException\s+[^\s\\]`
In master there are still +-25 consistency issues, ie. FQCN but without a root namespace. Shall we fix this?
Commits
-------
b074a71 Fixed expectedException annotations
This PR was merged into the 2.7 branch.
Discussion
----------
Security and validators messages translation to Latvian
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
Commits
-------
ff72f21 Security and validators messages translation to Latvian
This PR was merged into the 2.8 branch.
Discussion
----------
[VarDumper] Fix test
| Q | A
| ------------- | ---
| Branch? | 2.8
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Commits
-------
ae118e7 [VarDumper] Fix test
This PR was merged into the 2.7 branch.
Discussion
----------
[Form] FormView->isRendered() remove dead code and simplify the flow
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
The method `FormView->isRendered()` can be easily simplified by removing useless conditions and statements.
The condition at line 74 is useless because of the composite condition at line 70.
Removed dead code statement at line 84.
Commits
-------
40af42c [Form] FormView->isRendered() remove dead code and simplify the flow
This PR was merged into the 2.8 branch.
Discussion
----------
Use UUID for error codes for Form validator.
| Q | A
| ------------- | ---
| Branch? | 2.8
| Bug fix? | no
| New feature? | no
| BC breaks? | tiny, but it allowed http://symfony.com/doc/current/contributing/code/bc.html#changing-classes
| Deprecations? | no
| Tests pass? | I hope so :)
| Fixed tickets | -
| License | MIT
| Doc PR | -
Commits
-------
3693e83 Use UUID for error codes for Form validator.
* 2.7:
Revert "minor #20011 Use UUID for error codes for Form validator. (Koc)"
Use UUID for error codes for Form validator.
Fixed regression when exception message swallowed when logging it.
[HttpFoundation] Enable memcached tests with the latest memcached extension
This PR was submitted for the 2.8 branch but it was merged into the 2.7 branch instead (closes#20011).
Discussion
----------
Use UUID for error codes for Form validator.
| Q | A
| ------------- | ---
| Branch? | 2.8
| Bug fix? | no
| New feature? | no
| BC breaks? | tiny, but it allowed http://symfony.com/doc/current/contributing/code/bc.html#changing-classes
| Deprecations? | no
| Tests pass? | I hope so :)
| Fixed tickets | -
| License | MIT
| Doc PR | -
Commits
-------
6dbb186 Use UUID for error codes for Form validator.
This PR was merged into the 2.7 branch.
Discussion
----------
[DX] Fixed regression when exception message swallowed when logging it.
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | maybe, wait for Travis
| Fixed tickets | fixed regression introduced in #13418
| License | MIT
| Doc PR | -
The problem is: after merging referenced PR we cann't understand what error occured and why authentication request failed (see attached screenshot). Previously it displays exception message, but now it only displays class of the exception.
![9003644109](https://cloud.githubusercontent.com/assets/191082/18717550/f0df61c4-8028-11e6-8dbb-684e4928e913.jpg)
Commits
-------
042cb6b Fixed regression when exception message swallowed when logging it.
This PR was merged into the 2.7 branch.
Discussion
----------
[HttpFoundation] Enable memcached tests with the latest memcached extension
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
The check was introduced because of #9797. The issue is now fixed in the memcached extension, but only for the upcoming 3.0 release (for PHP7).
I've tested it with the latest memcached extension built from source and PHP 7.0.8.
Commits
-------
b482fb7 [HttpFoundation] Enable memcached tests with the latest memcached extension
This PR was merged into the 2.7 branch.
Discussion
----------
[TwigBridge] removed Twig null nodes (deprecated as of Twig 1.25)
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | yes (more about removing deprecation notices from newer versions of Twig)
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a (see twigphp/Twig#2123)
| License | MIT
| Doc PR | n/a
Commits
-------
12350c2 [TwigBridge] removed Twig null nodes (deprecated as of Twig 1.25)
This PR was merged into the 2.7 branch.
Discussion
----------
[Console] Fix parsing optionnal options with empty value in argv
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #19884
| License | MIT
If a command takes an option accepting an optional value, passing an empty value to this option will make it parsed as `null`, e.g:
`bin/console dummy --foo ""` gives `['foo' => null]`.
`bin/console dummy --foo=""` gives `['foo' => null]`.
Problems appear when adding an argument with a required value (let's call it `bar`):
`bin/console dummy --foo "" "bar-val"` gives `['foo' => null, 'bar' => 'bar-val']` which is OK.
But:
`bin/console dummy --foo="" "bar-val"`
> [RuntimeException]
Not enough arguments (missing: "bar").
The empty value is never considered, as `$argv` just return `"--foo="` for the option, the current implementation doesn't handle the empty value when using an equal as separator, so the `bar` argument value is considered as the `foo` one, giving a missing required argument at runtime.
This fixes it by explicitly considering the empty value if there is nothing immediately after the equal sign, so args/options correctly take their respective values.
Commits
-------
8952155 [Console] Fix empty optionnal options with = separator in argv