* 3.4: (22 commits)
use Precise on Travis to keep PHP LDAP support
Fix case sensitive sameSite cookie
[PropertyInfo] Use rawurlencode to escape PSR-6 keys
fix(security): ensure the 'route' index is set before attempting to use it
Fix registering lazy command services with autoconfigure enabled
Fix the design of the profiler exceptions when there is no message
[Config] Minor fix
document the TwigRenderer class deprecation
[Security] added more tests
[Security] fixed default target path when referer contains a query string
[Security] simplified tests
[Security] refactored tests
[WebProfilerBundle][TwigBundle] Fix infinite js loop on exception pages
[FrameworkBundle] fix ValidatorCacheWarmer: use serializing ArrayAdapter
Change "this" to "that" to avoid confusion
[VarDumper] Move locale sniffing to dump() time
[VarDumper] Use "C" locale when using "comma" flags
[Config] Make ClassExistenceResource throw on invalid parents
[DebugBundle] Added min_depth to Configuration
[Console] Add a factory command loader for standalone application with lazy-loading needs
...
* 3.3:
use Precise on Travis to keep PHP LDAP support
Fix case sensitive sameSite cookie
[PropertyInfo] Use rawurlencode to escape PSR-6 keys
fix(security): ensure the 'route' index is set before attempting to use it
Fix the design of the profiler exceptions when there is no message
[Config] Minor fix
[WebProfilerBundle] Fix full sized dump hovering in toolbar
* 3.2:
use Precise on Travis to keep PHP LDAP support
Fix case sensitive sameSite cookie
[PropertyInfo] Use rawurlencode to escape PSR-6 keys
fix(security): ensure the 'route' index is set before attempting to use it
[WebProfilerBundle] Fix full sized dump hovering in toolbar
* 2.8:
use Precise on Travis to keep PHP LDAP support
fix(security): ensure the 'route' index is set before attempting to use it
[WebProfilerBundle] Fix full sized dump hovering in toolbar
This PR was merged into the 2.7 branch.
Discussion
----------
use Precise on Travis to keep PHP LDAP support
| 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 |
Travis CI [started to roll out Ubuntu Trusty](https://blog.travis-ci.com/2017-07-11-trusty-as-default-linux-is-coming) as the default distribution. However, it seems that the PHP LDAP extension is missing on Trusty (see travis-ci/travis-ci#7067) starting to make our builds fail. Thus, I suggest to keep using Precise until the linked issue has been fixed.
Commits
-------
5441b1a use Precise on Travis to keep PHP LDAP support
This PR was merged into the 4.0-dev branch.
Discussion
----------
[Bridge\Doctrine] Fix BC with DI v3.4
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
As allowed by composer.json.
Code borrowed from branch 3.4
Commits
-------
b03f0bdb06 [Bridge\Doctrine] Fix BC with DI v3.4
This PR was merged into the 3.4 branch.
Discussion
----------
[Console] Fix registering lazy command services with autoconfigure enabled
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a
| License | MIT
| Doc PR | n/a
For
```yaml
_defaults:
autoconfigure: true
App\:
resource: '../../src/*'
App\Command\FooCommand:
tags:
- { name: console.command, command: foo }
```
Before you get the following error:
> Missing "command" attribute on tag "console.command" for service "App\Command\FooCommand"
Now the command is lazy.
----
Btw, @Tobion's https://github.com/symfony/symfony/pull/22734#discussion_r127410411
> Wouldn't it be more straightforward if aliases are just the additional tags using the command attribute as well?
Then there is no need for an alias property at all and this strange condition doesn't apply either.
Partially addressed here by removing the need for repeating the `command` attribute on each `console.command` tag
```yaml
# before
tags:
- { name: console.command, command: foo }
- { name: console.command, command: foo, alias: foobar }
# after
tags:
- { name: console.command, command: foo }
- { name: console.command, alias: foobar }
```
Tobias proposal:
```yaml
tags:
- { name: console.command, command: app:my-command }
- { name: console.command, command: app:my-alias }
```
I wanted to propose exactly the same at first, but finally found more clear to add a specific attribute for aliases, especially because relying on the order on which tags are defined sounds less good to me. Please tell me about your preference.
(And sorry for the noise around this feature, I want to polish it for 3.4)
Commits
-------
8a71aa31bb Fix registering lazy command services with autoconfigure enabled
This PR was merged into the 4.0-dev branch.
Discussion
----------
[Console] Make SymfonyQuestionHelper::ask optional by default
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | yes
| Tests pass? | yes (nothing in core depends on it)
| Fixed tickets | #... <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR | symfony/symfony-docs#... <!--highly recommended for new features-->
i noticed when writing commands i always keep doing
```php
$io = new SymfonyStyle($input, $output);
$answer = $io->ask('...', null, function ($value) { return $value; });
// instead of just
$answer = $io->ask('...');
```
only to bypass a built-in validation, of which im not sure why it's there. Note the base question helper doesnt make this assumption...
Commits
-------
2da429cd0a [Console] Make SymfonyQuestionHelper::ask optional by default
This PR was submitted for the master branch but it was merged into the 3.2 branch instead (closes#23586).
Discussion
----------
Fix case sensitive sameSite cookie
| Q | A
| ------------- | ---
| Branch? | 3.3
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #23585
| License | MIT
| Doc PR |
Commits
-------
14c310f5fb Fix case sensitive sameSite cookie
This PR was submitted for the 2.8 branch but it was merged into the 2.7 branch instead (closes#23238).
Discussion
----------
[Security] ensure the 'route' index is set before attempting to use it
| Q | A
| ------------- | ---
| Branch? | 2.8
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
```
// matching a request is more powerful than matching a URL path + context, so try that first
if ($this->urlMatcher instanceof RequestMatcherInterface) {
$parameters = $this->urlMatcher->matchRequest($request);
} else {
$parameters = $this->urlMatcher->match($request->getPathInfo());
}
return $path === $parameters['_route'];
```
Hi the issue here is the code is assuming a `_route` has been returned from the `match()` method.. however there is nothing to suggest that is always the case. For example if I just want to return a controller that is perhaps not added as an actual route I can & it works.. Although this will generate a notice warning.
**In terms of what happens if the `_route` is not defined should it return `false?` or actually perform a similar condition as `return $path === rawurldecode($request->getPathInfo());` **
I have an implementation of a router that is just returning a controller path and its arguments without a `_route` which works aside from this notice.
Commits
-------
7ae578cc1a fix(security): ensure the 'route' index is set before attempting to use it
This PR was merged into the 2.8 branch.
Discussion
----------
[WebProfilerBundle] Fix full sized dump hovering in toolbar
| Q | A
| ------------- | ---
| Branch? | 2.8 <!-- see comment below -->
| Bug fix? | yes
| New feature? | no <!-- don't forget updating src/**/CHANGELOG.md files -->
| BC breaks? | no
| Deprecations? | no <!-- don't forget updating UPGRADE-*.md files -->
| Tests pass? | yes
| Fixed tickets | #23563 <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR | N/A
Since #22953, the dump block shown on hovering the toolbar item takes the full width. But at least on OS X chrome, safari and firefox, the behavior is buggy and makes it unusable as the cursor can't reach the dumped content:
Honestly, I don't really understand the issue here and tried some tweaks until it works everywhere, (including trying to add a `.no-resize` class on `sf-toolbar-info` to avoid executing the related js event listener in case it was conflicting).
As shown in the screenshot, it also fixes the case where the dump wasn't full width under a certain size.
Commits
-------
28930c5 [WebProfilerBundle] Fix full sized dump hovering in toolbar
This PR was merged into the 3.3 branch.
Discussion
----------
[Config] Minor fix
| Q | A
| ------------- | ---
| Branch? | 3.3
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
My bad.
BTW @javiereguiluz I confirm my diagnosis on #23573
Here is the last exception in that error page:
Commits
-------
fa0b942 [Config] Minor fix
This PR was merged into the 3.4 branch.
Discussion
----------
[Console] Add a factory command loader for standalone application with lazy-loading needs
| Q | A
| ------------- | ---
| Branch? | 3.4 <!-- see comment below -->
| Bug fix? | no
| New feature? | yes <!-- don't forget updating src/**/CHANGELOG.md files -->
| BC breaks? | no
| Deprecations? | no <!-- don't forget updating UPGRADE-*.md files -->
| Tests pass? | yes (failure unrelated)
| Fixed tickets | https://github.com/symfony/symfony/pull/22734#issuecomment-314706630 <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR | todo (with https://github.com/symfony/symfony-docs/issues/8147)
So standalone applications can also benefit from the lazy loading feature without requiring a PSR-11 implementation specifically for this need.
The loader does not memoize any resolved command from factories, as it's the `Application` responsibility and the `ContainerCommandLoader` does not either (the PSR-11 does not enforce two successive calls to return the same value).
Commits
-------
9b40b4a [Console] Add a factory command loader for standalone application with lazy-loading needs
* 3.3:
[Security] added more tests
[Security] fixed default target path when referer contains a query string
[Security] simplified tests
[Security] refactored tests
[WebProfilerBundle][TwigBundle] Fix infinite js loop on exception pages
[FrameworkBundle] fix ValidatorCacheWarmer: use serializing ArrayAdapter
Change "this" to "that" to avoid confusion
[VarDumper] Move locale sniffing to dump() time
[VarDumper] Use "C" locale when using "comma" flags
[Config] Make ClassExistenceResource throw on invalid parents
This PR was squashed before being merged into the 2.7 branch (closes#23580).
Discussion
----------
Fix login redirect when referer contains a query string
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #19026, #23027, #23061, #23411, #23551
| License | MIT
| Doc PR | n/a
In 3.3, #19026 was merged to fix a bug that should have been fixed in 2.7. The fix was wrong anyway, so this PR fixes it the proper way.
The first two commits refactors test (using mocks for data objects is a bad idea and using too many mocks actually makes tests test nothing).
The actual fix is in the third commit.
Commits
-------
022ac0be09 [Security] added more tests
9c7a1406cb [Security] fixed default target path when referer contains a query string
b1f1ae26b4 [Security] simplified tests
3387612451 [Security] refactored tests
This PR was merged into the 3.4 branch.
Discussion
----------
[DebugBundle] Added min_depth to Configuration
This enables calling the recently-added `setMinDepth` function on `VarCloner`.
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | none
| License | MIT
| Doc PR | symfony/symfony-docs#8185
Commits
-------
30cd70d [DebugBundle] Added min_depth to Configuration
This PR was squashed before being merged into the 3.2 branch (closes#23558).
Discussion
----------
[FrameworkBundle] fix ValidatorCacheWarmer: use serializing ArrayAdapter
| Q | A
| ------------- | ---
| Branch? | 3.2
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | https://github.com/symfony/symfony/issues/23544
| License | MIT
| Doc PR | -
The `ValidatorCacheWarmer` was using an `ArrayAdapter` with `$storeSerialized=false`. This is a problem as inside the `LazyLoadingMetadataFactory` the metaData objects are mutated (parent class constraints are merged into it) after they have been written into the cache.
So this means that currently when warming up the validator cache actually the merged metaData version is finally taken from the `ArrayAdapter` and written into the `PhpFilesAdapter`.
Which then caused some duplicate constraints as the parent constraints are merged again after fetching from the cache inside `LazyLoadingMetadataFactory`.
This fix makes sure we serialize objects into the `ArrayAdapter`.
Writing a test case for this does not seem easy to me. Any ideas?
EDIT: Maybe its even safer to just clone the object when writing it into the cache?
```diff
diff --git a/src/Symfony/Component/Validator/Mapping/Factory/LazyLoadingMetadataFactory.php b/src/Symfony/Component/Validator/Mapping/Factory/LazyLoadingMetadataFactory.php
index 79ad1f2..88eaf33 100644
--- a/src/Symfony/Component/Validator/Mapping/Factory/LazyLoadingMetadataFactory.php
+++ b/src/Symfony/Component/Validator/Mapping/Factory/LazyLoadingMetadataFactory.php
@@ -117,7 +117,7 @@ class LazyLoadingMetadataFactory implements MetadataFactoryInterface
}
if (null !== $this->cache) {
- $this->cache->write($metadata);
+ $this->cache->write(clone $metadata);
}
```
Opinions?
Commits
-------
c0556cb204 [FrameworkBundle] fix ValidatorCacheWarmer: use serializing ArrayAdapter
This PR was merged into the 4.0-dev branch.
Discussion
----------
[DI] Optimize use of private and pre-defined services
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes (perf)
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
By making private services really private and taking into account that (un)setting pre-defined services is not allowed anymore, we can go one step further into optimizing the dumped container.
Commits
-------
c0c1881fe4 [DI] Optimize use of private and pre-defined services
This PR was merged into the 3.3 branch.
Discussion
----------
[Config] Make ClassExistenceResource throw on invalid parents
| Q | A
| ------------- | ---
| Branch? | 3.3
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #23564
| License | MIT
| Doc PR | -
Let's throw a more specific exception when a parent class/interface/trait is missing.
Fine tunes #23041
Commits
-------
53b01903ce [Config] Make ClassExistenceResource throw on invalid parents
This PR was merged into the 3.3 branch.
Discussion
----------
[VarDumper] Use "C" locale when using "comma" flags
| Q | A
| ------------- | ---
| Branch? | 3.3
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #23572
| License | MIT
| Doc PR | -
Adding trailing comma implicitly requires a C locale.
Commits
-------
305ae5e5d4 [VarDumper] Use "C" locale when using "comma" flags
This PR was merged into the 3.2 branch.
Discussion
----------
[DI] Change "this" to "that" in `findAndSortTaggedServices` doc to reduce confusion
Continuation of PR #23578 which I royally messed up.....
| Q | A
| ------------- | ---
| Branch? | 3.2
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| License | MIT
| Doc PR | None
I know this is extremely minor, but reading the description of this method, I got confused. Wondering if it's just me.
Where it says:
> The order of additions must be respected for services having the same priority, and knowing that the \SplPriorityQueue class does not respect the FIFO method, we should not use **this** class.
Should it not say "we should not use **that** class"?
Commits
-------
04b7b04b65 Change "this" to "that" to avoid confusion
Nicolas Grekas
Christian Flothmann
Fabien Potencier
Mike Francis
Kévin Dunglas
Gavin Staniforth
Javier Eguiluz
Maxime Steinhausser
Robin Chalas
Maxime Steinhausser
David Maicher
Shaun Simmons
