This PR was merged into the 2.7 branch.
Discussion
----------
[HttpFoundation] fixed return type of method HeaderBag::get
| Q | A
| ------------- | ---
| Branch? | 3.3
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| License | MIT
<!--
- Bug fixes must be submitted against the lowest branch where they apply
(lowest branches are regularly merged to upper ones so they get the fixes too).
- Features and deprecations must be submitted against the master branch.
- Please fill in this template according to the PR you're about to submit.
- Replace this comment by a description of what your PR is solving.
-->
Commits
-------
6b608e8 [HttpFoundation] fixed return type of method HeaderBag::get
This PR was merged into the 2.7 branch.
Discussion
----------
[HttpFoundation] Added "resource" type on Request::create docblock
| 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 | -
Replaces #24902
Commits
-------
d1a96ca [HttpFoundation] Added "resource" type on Request::create docblock
This PR was merged into the 4.0 branch.
Discussion
----------
[Console] Fix using finally where the catch can also fail
| Q | A
| ------------- | ---
| Branch? | 4.0
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #25872
| License | MIT
| Doc PR | -
Restores code to 3.4 state, which is free from the issue.
Commits
-------
763a914 [Console] Fix using finally where the catch can also fail
This PR was merged into the 2.7 branch.
Discussion
----------
[Validator] add missing parent isset and add test
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | yes
| New feature? | no <!-- don't forget to update src/**/CHANGELOG.md files -->
| BC breaks? | no
| Deprecations? | no <!-- don't forget to update UPGRADE-*.md files -->
| Tests pass? | yes
| Fixed tickets | #25843 <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR |
This is a little PR to fix the error and to add the parent::__isset to the Constraints class, I'm adding a test too to be sure that this is not possible again.
Btw, since this is instagihub, I'm coding at home and then I see a squirrel :
![img_3187 heic](https://user-images.githubusercontent.com/3451634/35149517-a9ddc0ba-fd16-11e7-9e22-cdba0b677071.jpeg)
Commits
-------
bcb79a3 [Validator] add missing parent isset and add test
This PR was merged into the 3.3 branch.
Discussion
----------
[Process] Skip environment variables with false value in Process
| Q | A
| ------------- | ---
| Branch? | master, 4.0, 3.3, 3.4
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | ~
| License | MIT
| Doc PR | ~
With the commit 03adce239d, all env variables are injecting in the process, and the env variable with the `false` value are converted in empty string.
For example, it's problematic with the bundle SymfonyWebServerBundle and the last version of the [symfony/framework-bundle recipe](https://github.com/symfony/recipes/blob/master/symfony/framework-bundle/3.3/public/index.php#L11), because the WebServer override the value of `APP_ENV` with `false` to override previously loaded variables (c5a1218555), and with the commit 03adce239d, the `APP_ENV` variable is injected with a empty string value, instead of not being injected, as was the case before.
This PR not use the same logic as the [getDefaultEnv()](https://github.com/symfony/symfony/blob/4.0/src/Symfony/Component/Process/Process.php#L1553) method.
Commits
-------
2daf4f9 [Process] Skip environment variables with false value in Process
This PR was merged into the 4.1-dev branch.
Discussion
----------
[Config] allow changing the path separator
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #22268
| License | MIT
| Doc PR |
In bundles configuration, no dots are allowed in key names, but as a standalone config, there is no such limitation that should be enforced.
By using current `NodeBuilder` it should be possible to change path separator used (for example: `/`)
Commits
-------
15e05e1 Add feature allowing change of Path Separator
This PR was merged into the 4.0 branch.
Discussion
----------
[Yaml] trim spaces from unquoted scalar values
| Q | A
| ------------- | ---
| Branch? | 4.0
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #25856
| License | MIT
| Doc PR |
Commits
-------
5e70ee898e trim spaces from unquoted scalar values
This PR was merged into the 2.8 branch.
Discussion
----------
Formatting fix in upgrade 3.0 document
| Q | A
| ------------- | ---
| Branch? | 2.8
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | no
| License | MIT
| Doc PR | no
Fixes formatting in upgrade file.
This fix was already done in https://github.com/symfony/symfony/pull/22436 but applied to 3.2.
Formatting error still exists in 2.8, 3.0, 3.1. I don't know if this was intentional.
Commits
-------
dc6d5e1083 Formatting fix in upgrade 3.0 document
This PR was merged into the 2.7 branch.
Discussion
----------
do not conflict with egulias/email-validator 2.0+
| Q | A
| ------------- | ---
| Branch? | 2.7
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | https://github.com/symfony/symfony/pull/25851#issuecomment-359166972
| License | MIT
| Doc PR |
Not allowing `egulias/email-validator` 2.0+ will prevent applications using this package from updating to the next patch release.
Commits
-------
1a1aaa74d6 Revert "bug #25851 [Validator] Conflict with egulias/email-validator 2.0 (emodric)"
This PR was merged into the 2.7 branch.
Discussion
----------
[Validator] Conflict with egulias/email-validator 2.0
| Q | A
| ------------- | ---
| Branch? | 2.7 (and 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
For code that uses `EmailValidator` with strict email checking using `egulias/email-validator` and that aims to support both Symfony 2.8 LTS and Symfony 3.4/4.0, it is impossible to have a working set of dependencies with Symfony 2.7/Symfony 2.8 if one wishes to use `egulias/email-validator: ^1.2|^2.1`.
This is because Composer happily installs `egulias/email-validator` latest 2.1.x which is not compatible with Symfony 2.8.
This would prevent installation of v2.1.x on Symfony 2.7/2.8.
Commits
-------
72d8e8adb0 [Validator] Conflict with egulias/email-validator 2.0
This PR was squashed before being merged into the 4.1-dev branch (closes#25493).
Discussion
----------
[Serializer] `default_constructor_arguments` context option for denormalization
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | (there is no RFC for this)
| License | MIT
| Doc PR | https://github.com/symfony/symfony-docs/pull/8914
## Problems
In the case you want to deserialize value-objects, if all the data required by its constructor are **not** given as input, the serializer will throw a simple `RuntimeException` exception. This makes impossible to catch it. (as current fix on my projects I use exception message to be sure to catch the good one X.x")
The second problem is a missing feature to fill the required object with an empty one. This needs to be defined by the user because the serializer can't guess how to build it.
Here is a project that exposes the problem of the current behavior. https://github.com/Nek-/api-platform-value-object
## Solutions suggested
I suggest a solution in 2 parts because the second part is more touchy.
1. Replace the current exception by a new specific one
2. Add a new `empty_data` option to the context of deserialization so you can specify data for objects impossible to instantiate, this is great because the serializer no more throw exception and the validator can then work as expected and send violations to the user. This solution is inspired by forms solutions to fix the issue with value objects
Here is what you can do with this feature:
```php
class DummyValueObject
{
public function __construct($foo, $bar) { $this->foo = $foo; $this->bar = $bar; }
}
$empty = new DummyValueObject('', '');
$result = $normalizer->denormalize(['foo' => 'Hello'], DummyValueObject::class, 'json', [
'empty_data' => [
DummyValueObject::class => $empty,
],
]);
// It's impossible to construct a DummyValueObject with only "foo" value. So the serializer
// will replace it with the given empty data
```
There are 2 commits so I can quickly provide you only the first point if you want. Hope you'll like this.
## Solution after discussion
1. New exception `MissingConstructorArgumentsException`
2. New context option `default_constructor_arguments`
```php
class DummyValueObject
{
public function __construct($foo, $bar) { $this->foo = $foo; $this->bar = $bar; }
}
$result = $normalizer->denormalize(['foo' => 'Hello'], DummyValueObject::class, 'json', [
'default_constructor_arguments' => [
DummyValueObject::class => ['foo' => '', 'bar' => ''],
],
]);
// DummyValueObject is contructed with the given `foo` and empty `bar`
```
Commits
-------
1523a85542 [Serializer] context option for denormalization
This PR was merged into the 3.4 branch.
Discussion
----------
[Serializer] Fix undefined class
| 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
bad merge
Commits
-------
bc7b5d2 [Serializer] fix undefined class
This PR was merged into the 3.3 branch.
Discussion
----------
[FrameworkBundle] fix DI extension tests
| 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 |
The failing tests relied on the assets integration being enabled. Since
we never did enable this explicitly, the assets related definitions are
removed now that #25789 was merged which fixed#25760.
Commits
-------
1cb8f69 [FrameworkBundle] fix DI extension tests
The failing tests relied on the assets integration being enabled. Since
we never did enable this explicitly, the assets related definitions are
removed now that #25789 was merged which fixed#25760.