This PR was merged into the 4.4 branch.
Discussion
----------
[SecurityGuard] Deprecate returning non-boolean values from checkCredentials()
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | yes
| Tests pass? | yes
| Fixed tickets | prepares #33228
| License | MIT
| Doc PR | TODO
This PR suggests to deprecate implementations of `Symfony\Component\Security\Guard\AuthenticatorInterface::checkCredentials()` that return non-boolean values. This will allow us to add `bool` as return type declaration in the future.
https://github.com/nicolas-grekas/symfony/pull/26#discussion_r317084918
/cc @nicolas-grekas @weaverryan
Commits
-------
a0ca3afeca Deprecate returning non-boolean values from checkCredentials().
This PR was merged into the 4.4 branch.
Discussion
----------
[PhpUnitBridge] Bump SYMFONY_PHPUNIT_VERSION
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | yes
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets | #... <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR | symfony/symfony-docs#... <!-- required for new features -->
👋 here we are again :)
i was wondering... cant we just install the latest minor by default, thus `SYMFONY_PHPUNIT_VERSION=8`? There should be no BC breaks, and would reduce the maintenance to one major bump, once a year ... isnt it? 🤔
Commits
-------
9cbd0e8809 [PhpUnitBridge] Bump SYMFONY_PHPUNIT_VERSION
This PR was merged into the 4.4 branch.
Discussion
----------
Add missing return annotations on magic methods
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
My script had a bug, spotted by reviewing #33267
These annotations express our intention to add real return types in a future major release (likely v6)
Commits
-------
10983fcb17 Add missing return annotations on magic methods
* 4.3:
Revert "bug #31620 [FrameworkBundle] Inform the user when save_path will be ignored (gnat42)"
[Form][PropertyPathMapper] Avoid extra call to get config
[HttpKernel] remove unused fixtures
* 3.4:
Revert "bug #31620 [FrameworkBundle] Inform the user when save_path will be ignored (gnat42)"
[Form][PropertyPathMapper] Avoid extra call to get config
This PR was merged into the 4.3 branch.
Discussion
----------
[HttpKernel] remove unused fixtures
| Q | A
| ------------- | ---
| Branch? | 4.3
| Bug fix? | no
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets | <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR |
those are remnants of bundle inheritance that has been removed in sf 4
Commits
-------
5ffec16396 [HttpKernel] remove unused fixtures
This PR was merged into the 3.4 branch.
Discussion
----------
[Form][PropertyPathMapper] Avoid extra call to get config
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Commits
-------
3e2aada2d8 [Form][PropertyPathMapper] Avoid extra call to get config
This PR was merged into the 4.4 branch.
Discussion
----------
Add return types to internal & magic methods when possible
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #33236#33228
| License | MIT
| Doc PR | -
Commits
-------
962dcfeed0 Add return types to internal & magic methods when possible
This PR was merged into the 4.4 branch.
Discussion
----------
[Mime] Add Address::fromString
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #33086
| License | MIT
| Doc PR | https://github.com/symfony/symfony-docs/pull/12128
This will allow to create a Address from a string such as 'Name <name@example.com>'
Example:
```php
$address = Address::fromString("Name <name@example.com>");
```
Commits
-------
75ea8d0d67 Add Address::fromString
This PR was merged into the 4.4 branch.
Discussion
----------
Mark all dispatched event classes as final
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets |
| License | MIT
| Doc PR |
I think we should mark all our Event classes as final. There is no point in people extending them as the libraries that use the event, will only dispatch this event. So extending events in user-land achieves nothing as the subclasses won't be dispatched.
I'm not talking about the base events that are meant to be extended like KernelEvent, but the leaf events like ExceptionEvent, ResponseEvent etc.
Then we can also make them real final in 5.0 as the events are value objects that should not be mocked.
Commits
-------
4bb38eec89 Mark all dispatched event classes as final
This PR was merged into the 4.4 branch.
Discussion
----------
fix deprecated call to setLocale with null
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets | <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR |
Fix a potentially invalid call since #33272
Commits
-------
26f9afe8d1 fix deprecated call to setLocale with null
This PR was merged into the 4.4 branch.
Discussion
----------
[HttpKernel] deprecate global dir to load resources from
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | yes <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets | #31915 <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR |
Replaces #31958
Here two example deprecations by adding files in the deprecated locations:
```
Overwriting the resource "@AcmeBundle/Resources/config/routing.yaml" with "/vagrant/src/Resources/AcmeBundle/config/routing.yaml" is deprecated since Symfony 4.4 and will be removed in 5.0.
Loading the file "foobar.yaml" from the global resource directory "/vagrant/src" is deprecated since Symfony 4.4 and will be removed in 5.0.
```
Commits
-------
aa82566f76 [HttpKernel] deprecate global dir to load resources from
* 4.3:
Do not extend the new SF 4.3 ControllerEvent so we can make it final
Backported return type violation bugfixes.
[FrameworkBundle] Fix BrowserKit assertions to make them compatible with Panther
This PR was merged into the 4.3 branch.
Discussion
----------
[HttpKernel] Do not extend the new SF 4.3 ControllerEvent so we can make it final
| Q | A
| ------------- | ---
| Branch? | 4.3
| Bug fix? | no
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| BC breaks? | unlikely
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets |
| License | MIT
| Doc PR |
See https://github.com/symfony/symfony/pull/33152#discussion_r313846346
Remember the ControllerEvent is new in SF 4.3 so we just go back to what it was before 4.3
Commits
-------
00140b6a7c Do not extend the new SF 4.3 ControllerEvent so we can make it final
This PR was merged into the 4.4 branch.
Discussion
----------
Removed calls to Twig\Environment::loadTemplate()
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | N/A
| License | MIT
| Doc PR | N/A
This PR prepares #33039. Twig 3 does not have the `loadTemplate()` anymore, so this PR replaces calls to that method.
Commits
-------
ea9e375b0b Removed calls to Twig\Environment::loadTemplate().
This PR was merged into the 4.4 branch.
Discussion
----------
[Intl] make polyfill classes abstract, fix edge case
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
While working on return types, I keep stumbling on those classes that are `@internal`, but for which we must not add return types, because they're polyfills and are extended by stubs found in the `Resources` folder of the component.
Making the polyfills abstract fixes the linting issue.
This made me discover we have a glitch in the `getPattern()` implementation, that makes our version diverge from intl's. Fixed here too.
On 4.4 because let's not disrupt 3.4.
Commits
-------
c757b95aed [Intl] make polyfill classes abstract, fix edge case
This PR was merged into the 4.4 branch.
Discussion
----------
[Translation] deprecate support for null locales
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | yes
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
Commits
-------
e6b6a9d33a deprecate support for null locales
This PR was squashed before being merged into the 4.3 branch (closes#33216).
Discussion
----------
[Mime] Trim and remove line breaks from NamedAddress name arg
| Q | A
| ------------- | ---
| Branch? | 4.3
| Bug fix? | YES
| New feature? | NO
| BC breaks? | NO <!-- see https://symfony.com/bc -->
| Deprecations? | NO <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | YES <!-- please add some, will be required by reviewers -->
| Fixed tickets | N/A <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR | N/A
This patch trims the name argument of named address in order to avoid some cases where the name is a non-empty input consisting of whitespace and line breaks which would lead to forming of addresses such as `" " <mail@example.org>` <sup>1</sup>
---
In a large Symfony codebase that deals with sending large volumes of email we encountered an issue after the Mailer was changed from PHPMailer to the Symfony Mailer component.
The issue: Some emails would not render correctly as either plaintext or html but instead the original email source with headers and quoted-printable content would render in clients such as MS Outlook 2003, later versions of Outlook and other clients do not seem affected.
After some investigation we found that the error came from a line that looked like this: `$message->addTo(new NamedAddress($contact->getEmailCanonical(), $contact->getFullName()))`.
The `getFullName` method simply concated the first/last name with a space in between. For contacts without either, this resulted in representation 1. This causes MS Outlook 2003 (potentially Outlook 2000 as well but this was not tested) to malfunction and completely fail to render the email.
This patch aims to fix the aforementioned issue.
Commits
-------
e491e3a594 [Mime] Trim and remove line breaks from NamedAddress name arg
This PR was merged into the 4.4 branch.
Discussion
----------
[TwigBridge] Mark all classes extending twig as @final
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes-ish
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets | refs #33039
| License | MIT
| Doc PR | n/a
Classes defining extensions/nodes/node visitors/token parsers should not be changed. They should be final.
That would also help with Twig 3.0 which introduces type hints (including return types).
Commits
-------
d657459a5f [TwigBridge] Mark all classes extending twig as @final
This PR was merged into the 4.4 branch.
Discussion
----------
[Messenger] remove patch release BC layer of durable and expiring delay
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets |
| License | MIT
| Doc PR |
Removes the small BC layer of #33127 from 4.3 in 4.4
Commits
-------
d5aaf44529 [Messenger] remove patch release BC layer of durable and expiring delay
This PR was merged into the 4.4 branch.
Discussion
----------
[4.4] Add return types on internal|final|private methods (bis)
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Found while working on #33259
`: self` is used for final methods only. I'd have preferred using `: object` but that's not possible on PHP 7.1
Commits
-------
23faee406f [4.4] Add return types on internal|final|private methods (bis)
This PR was merged into the 4.4 branch.
Discussion
----------
Add types to routing and DI configuration traits.
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no
| BC breaks? | I don't think so.
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #32179, #33228
| License | MIT
| Doc PR | N/A
This PR backports the type declarations added to the configurator traits of the Routing and DI components. These traits expose only final methods, so it should be pretty safe to add return types to them.
The only scenario I could make up where this change will break something is if a trait is used to override a method: https://3v4l.org/EAsk8 But I doubt that those traits are used that way.
On master, we've used the `object` return type for the fluent methods. That type is not available on 4.4 where we have to support php 7.1. I'm using `self` instead. Since the methods are final and thus cannot be overridden, I believe that we shouldn't run into covariance issues here, so `self` should be safe.
Commits
-------
1ca30c97e6 Add types to roting and DI configuration traits.
This PR was merged into the 4.4 branch.
Discussion
----------
[ErrorHandler] Registering basic exception handler to handle early failures
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
This behavior was previously handled by the removed `ExceptionHandler` class in https://github.com/symfony/symfony/pull/32637.
As this method is mainly called during Kernel boot, where nothing is yet available, the Response content is always HTML. Otherwise, If all goes well on booting, this exception handler will be replaced in `DebugHandlersListener` class:
8073b8abfb/src/Symfony/Component/HttpKernel/EventListener/DebugHandlersListener.php (L139)
where the advanced exception handler mechanism is activated.
Commits
-------
a2077a2369 Registers basic exception handler to handle early failures
This PR was merged into the 4.3 branch.
Discussion
----------
Revert "bug #33092 [DependencyInjection] Improve an exception message"
| Q | A
| ------------- | ---
| Branch? | 4.3
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
As reminded by @ro0NL in https://github.com/symfony/symfony/pull/33092#issuecomment-520138148, it looks like we forgot that `CheckDefinitionValidityPass` already checks and suggests for leading slashes.
Why didn't you get the exception from `CheckDefinitionValidityPass` @fabpot?
Commits
-------
ed590ca16b Revert "bug #33092 [DependencyInjection] Improve an exception message (fabpot)"
This PR was squashed before being merged into the 4.3 branch (closes#33124).
Discussion
----------
[Config] Add handling for ignored keys in ArrayNode::mergeValues.
| Q | A
| ------------- | ---
| Branch? | 4.3
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #33101
| License | MIT
| Doc PR | N/A
Fix case where normalized data accepting and keeping ignored keys is rejected during merge.
Commits
-------
311e1c4e9e [Config] Add handling for ignored keys in ArrayNode::mergeValues.
This PR was merged into the 3.4 branch.
Discussion
----------
[Router] Fix TraceableUrlMatcher behaviour with trailing slash
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets | #32149
| License | MIT
| Doc PR | ¤
<!--
Replace this notice by a short README for your feature/bugfix. This will help people
understand your PR and can be used as a start for the documentation.
Additionally (see https://symfony.com/roadmap):
- Bug fixes must be submitted against the lowest maintained 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 branch 4.4.
- Legacy code removals go to the master branch.
-->
This pull requests fixes the bug #32149. This issue was about TraceableUrlMatcher having a wrong behaviour regarding trailing slashes (according to UrlMatcher and documentation).
With this pull requests, the test class TraceableUrlMatcherTest now extends UrlMatcherTest, to prevent such behaviour digression.
Thanks @nicolas-grekas for his feedback on the issue #32149
Commits
-------
fd1cb443fd [Router] Fix TraceableUrlMatcher behaviour with trailing slash
This PR was merged into the 3.4 branch.
Discussion
----------
[Security/Core] UserInterface::getPassword() can return null
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Our very own `User` class can return null already.
Commits
-------
00d7f8cde7 [Security/Core] UserInterface::getPassword() can return null
This PR was merged into the 3.4 branch.
Discussion
----------
Fix inconsistent return points
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #17201 in preparation for #33228
| License | MIT
| Doc PR | N/A
Inconsistent return points in methods prevent adding return types. I thought, I'll give it a try and fix them. After this PR, PhpStorm's inspection still finds 39 issues, but as far as I can tell, they're either false positives or fixture code.
Commits
-------
f5b6ee9de1 Fix inconsistent return points.
This PR was merged into the 4.4 branch.
Discussion
----------
[DI] add return-types to generated containers
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
A chunk of #33236 ready for 4.4
Commits
-------
917091955c [DI] add return-types to generated containers
This PR was merged into the 4.4 branch.
Discussion
----------
[Ldap] Make LdapUser implement EquatableInterface
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Bugfix because it is required for password migrations https://github.com/symfony/symfony/blob/4.4/src/Symfony/Component/Ldap/Security/LdapUserProvider.php#L128
Commits
-------
ae255095ea [Ldap] Make LdapUser implement EquatableInterface
This PR was merged into the 4.3 branch.
Discussion
----------
Fix handling for session parameters
| Q | A
| ------------- | ---
| Branch? | 4.3
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| License | MIT
Commits
-------
b8c9e40980 Fix handling for session parameters
This PR was merged into the 4.4 branch.
Discussion
----------
[Mailer] Remove the auth mode DSN option and support in the eSMTP transport
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | yes
| BC breaks? | yes <!-- see https://symfony.com/bc -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets | n/a
| License | MIT
| Doc PR | -
The authentication mode can be automatically negotiated between the Mailer and the SMTP server. There is an option to force it to a given auth mode, but I don't see any valid use case. So, let's remove that feature.
Commits
-------
34cbda53c4 [Mailer] removed the auth mode DSN option and support in the eSMTP transport
* 4.3:
[Routing] Add a param annotation for $annot.
[DI] fix docblock
[Console] fix docblock
Add missing translations for Armenian locale
[Process] Added missing return type.
[Process] Doc block backport.
Added doc block for Registry::supports().
[Cache] Fix predis test
Don't duplicate addresses in Sendgrid Transport
Remove unnecessary statement
Fix some docblocks.
[Messenger] make delay exchange and queues durable like the normal ones by default
Cancel delayed message if handler fails
Added tests for #32370
* 3.4:
[Routing] Add a param annotation for $annot.
[DI] fix docblock
Add missing translations for Armenian locale
[Process] Doc block backport.
Fix some docblocks.
This PR was merged into the 4.4 branch.
Discussion
----------
[Monolog] Added ElasticsearchLogstashHandler
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
This PR was initially [submitted on Monolog](https://github.com/Seldaek/monolog/pull/1334).
It has been refused , but Jordi suggested to add it to the symfony bridge. So here we go :)
---
ATM, there are few options to push log to Elastic Stack in order to play with them:
* install logstash and use a Gelf handler. It works but you have to install logstash and configure it. Not an easy task. More over, it need an extra PHP package
* use the ES handler: It does not play well with context and extra: Kibana is not able to filter on nested object. And this handler is tightly coupled to the ElasticaFormatter formater. More over, it need an extra PHP package
* use something to parse file logs. This is really a bad idea since it involves a parsing... More over a daemon is needed to do that (file beat / logstash / you name it)
This is why I'm introducing a new Handler.
* There is not need to install anything (expect ES, of course)
* It play very well with Kibana, as it uses the Logstash format
* It requires symfony/http-client, but in a modern PHP application (SF 4.3) this dependency is already present
* It slow down a bit the application since it trigger an HTTP request for each logs. But symfony/http-client is non-blocking. If you want to use it in production, I recommend to wrap this handler in a buffer handler or a cross-finger handle to have only one HTTP call.
---
Some performance consideration en a prod env with a buffer handler + this one
* with push to ES: https://blackfire.io/profiles/f94ccf35-9f9d-4df1-bfc5-7fa75a535628/graph
* with push to ES commented: https://blackfire.io/profiles/6b66bc18-6b90-4341-963f-797f7a7a689c/graph
As you can see, as requests are made synchronously, there is no penalty on `AppKernel::Handler()` 😍! But the PHP worker has more work to do, and it's busy much more time (about X2)
I explained everything in the PHP Doc Block
---
This is what you can expect **out of the box**
Commits
-------
1587e9a4e2 [Monolog] Added ElasticsearchLogstashHandler
This PR was squashed before being merged into the 4.4 branch (closes#33212).
Discussion
----------
[VarDumper] Add test dump image
| Q | A
| ------------- | ---
| Branch? | 4.4 <!-- see below -->
| Bug fix? | no
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | <!-- please add some, will be required by reviewers -->
| Fixed tickets | #33107. <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR | <!-- required for new features -->
<!--
Replace this notice by a short README for your feature/bugfix. This will help people
understand your PR and can be used as a start for the documentation.
Additionally (see https://symfony.com/roadmap):
- Bug fixes must be submitted against the lowest maintained 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 branch 4.4.
- Legacy code removals go to the master branch.
-->
Commits
-------
8393a9b5c1 [VarDumper] Add test dump image
This PR was merged into the 4.4 branch.
Discussion
----------
Add types to private and final methods
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #32179
| License | MIT
| Doc PR | N/A
Backports from #33154.
Commits
-------
1b880677d4 Add types to private and final methods.
This PR was merged into the 4.3 branch.
Discussion
----------
[Messenger] DispatchAfterCurrentBusMiddleware does not cancel messages from delayed handlers
| Q | A
| ------------- | ---
| Branch? | 4.3 for bug fixes
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? |no
| Tests pass? | yes (thanks @Nyholm)
| Fixed tickets | #32370
| License | MIT
| Doc PR | -
This is a fix for #32370. There is no need for anything sophisticated. There is no recursion or fancy stuff going on, just a queue of message handled sequentially. A simple variable is enough to keep track of the queue state.
Thanks @Nyholm for the test, it would clearly have been the hardest part of the job.
Commits
-------
1f5c8a6790 Cancel delayed message if handler fails
35c76a385d Added tests for #32370
This PR was merged into the 4.4 branch.
Discussion
----------
[Messenger] InMemoryTransport handle acknowledged and rejected messages
| Q | A
| ------------- | ---
| Branch? | 4.4 <!-- see below -->
| Bug fix? | no
| New feature? | yes <!-- please update src/**/CHANGELOG.md files -->
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets | #... <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR | symfony/symfony-docs/pull/12045 <!-- required for new features -->
This PR do 2 things:
* Limit receiving messages from InMemoryTransport to 1 (reverted, another PR will fix the bug: worker does not stop when receiver return more messages than the number specify by the `--limit` option)
* Handle acknowledged and rejected messages in InMemoryTransport. Currently, it does not care about acknowledged and rejected messages. So it always return all messages that have been sent. So if we run console command `messenger:consume`, it will never stop, even though we set the `--limit` option.
For more information, please check the [reproduction](https://github.com/tienvx/symfony-messenger-in-memory-reproduction) project for the expected behavior.
See also my [messenger-memory-transport](https://github.com/tienvx/messenger-memory-transport) project
Commits
-------
71e7bdff22 [Messenger] InMemoryTransport handle acknowledged and rejected messages
This PR was merged into the 4.3 branch.
Discussion
----------
[Messenger] make delay exchange and queues durable like the normal ones by default
| Q | A
| ------------- | ---
| Branch? | 4.3
| Bug fix? | yes
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets | #32891
| License | MIT
| Doc PR |
This also imrproves BC of #32631 by only adding the new expiry argument in case the delay queue name was not overwritten using the options. I will remove the checks in 4.4 again. Please merge this PR before releasing the new 4.3 version so that 32631 and this PR are part of one release.
Commits
-------
e5ecda6de1 [Messenger] make delay exchange and queues durable like the normal ones by default
This PR was merged into the 4.4 branch.
Discussion
----------
[Intl] Full alpha3 language support
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets | #33136
| License | MIT
| Doc PR | symfony/symfony-docs#... <!-- required for new features -->
I'll validate some more cases with tests.
Commits
-------
29aee2ddf2 [Intl] Full alpha3 language support
This PR was merged into the 4.4 branch.
Discussion
----------
[ErrorHandler] Added call() method utility to turns any PHP error into \ErrorException
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | https://github.com/symfony/symfony/issues/32936
| License | MIT
| Doc PR | symfony/symfony-docs#...
**Issue**
There is no easy way to catch PHP warnings, though some progress has been made in this area for PHP 8.0 (https://wiki.php.net/rfc/consistent_type_errors).
**Before**
```php
$file = file_get_contents('unknown.txt');
// PHP Warning: file_get_contents(unknown.txt): failed to open stream: No such file or directory
// workaround:
$file = @file_get_contents('unknown.txt');
if (false === $file) {
$e = error_get_last();
throw new \ErrorException($e['message'], 0, $e['type'], $e['file'], $e['line']);
}
```
**After**
```php
$file = ErrorHandler::call('file_get_contents', 'unknown.txt');
// or
$file = ErrorHandler::call(static function () {
return file_get_contents('unknown.txt');
});
// or (PHP 7.4)
$file = ErrorHandler::call(fn () => file_get_contents('unknown.txt'));
```
All credits to @nicolas-grekas https://github.com/symfony/symfony/issues/32936#issuecomment-518152681 and @vudaltsov for the idea.
Commits
-------
0faa855b5e Added ErrorHandler::call() method utility to turns any PHP warnings into `\ErrorException`
This PR was merged into the 4.4 branch.
Discussion
----------
[Intl] Validate country preferred alpha code mapping
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets | #... <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR | symfony/symfony-docs#... <!-- required for new features -->
Use the same approach for the preferred alpha2-to-alpha3 code mapping, like Languages does. It provides us some more support in terms of maintenance.
The added `CD => COD` was previously implied based on ordering upstream.
Commits
-------
345305904b [Intl] Validate region preferred alpha code mapping
This PR was merged into the 4.3 branch.
Discussion
----------
[Mailer] Don't duplicate addresses in Sendgrid Transport
| Q | A
| ------------- | ---
| Branch? | 4.3
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #33135
| License | MIT
| Doc PR | N/A
Sendgrid requires the `to`, `cc` and `bcc` fields to be unique
Commits
-------
2706a9763f Don't duplicate addresses in Sendgrid Transport
The casting of `$id` to string inside the second foreach loop in `\Symfony\Component\DependencyInjection\Dumper\PhpDumper::addMethodMap` is redundant, as the variable is not used after the casting inside nor outside the loop (while still in the loop, it gets overriden upon next iteration).
Fixes#33206
This PR was merged into the 3.4 branch.
Discussion
----------
[Console] fixed a PHP notice when there is no function in the stack trace of an Exception
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| BC breaks? | no <!-- see https://symfony.com/bc -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tests pass? | yes <!-- please add some, will be required by reviewers -->
| Fixed tickets | n/a
| License | MIT
| Doc PR | n/a
Under certain circumstances, the `function` is not present in the stack trace. That's the case for instance when an error occurs on a line like this one `return require "somefile";`.
Commits
-------
ddb47358ae [Console] fixed a PHP notice when there is no function
This PR was merged into the 4.4 branch.
Discussion
----------
[ErrorHandler] trigger deprecation in DebugClassLoader when child class misses a return type
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | https://github.com/symfony/symfony/issues/30123
| License | MIT
| Doc PR | TODO
I wanted to push something to show the advancement and get feedback.
I pushed two versions : one with dedicated functions for code clarity (DebugClassLoader.php) and one withtout (DebugClassLoader___.php). It would be nice if some people with Blackfire could compare the performances.
So let's be clear, we are never gonna be able to cover all cases! We can however cover the vast majority.
Current non covered cases and problems :
- We assume that if there is more than 2 returned types, we cannot do anything. Even if it could technically be possible.
- We assume that any returned type that doesn't fit our "returnable" types list is a class. We don't check at all if this class actualy exists.
- We don't handle spaces in types. The types stop at the first space.
- That means we don't handle (yet) the callable type with spaces (cf https://github.com/symfony/symfony/issues/29969)
- Vendor code extending other vendor core triggers the deprecations 😕
Commits
-------
aa338c8b42 Import return annotations from vendors
10fc13e4a7 [ErrorHandler] Handle return types in DebugClassLoader
