This PR was squashed before being merged into the 3.4 branch.
Discussion
----------
[Console] Fix OutputStream for PHP 7.4
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#36166
| License | MIT
From PHP 7.4, `fwrite` function now returns false for any failure: https://www.php.net/manual/en/migration74.incompatible.php#migration74.incompatible.core.fread-fwrite
Actually, the note in the PHP documentation is not exact: for PHP 7.3 and lower, `fwrite` function did return false when arguments passed in to the function were invalid, and 0 for other failures. From PHP 7.4, it returns false for any failure.
We can see it in the source code: for PHP 7.3: a1a8d14485/ext/standard/file.c (L1140)
Compare to PHP 7.4: https://github.com/php/php-src/blob/master/ext/standard/file.c#L1136
I update `OutputStream::doWrite()` to keep the same behavior as before.
Commits
-------
b375f93ed7 [Console] Fix OutputStream for PHP 7.4
This PR was squashed before being merged into the 4.4 branch.
Discussion
----------
add missing gitattributes for phpunit-bridge
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tickets |
| License | MIT
| Doc PR |
Seems like the phpunit bridge has been forgotten in https://github.com/symfony/symfony/pull/33579
Commits
-------
d4c052a2fa add missing gitattributes for phpunit-bridge
This PR was submitted for the master branch but it was merged into the 3.4 branch instead.
Discussion
----------
[Validator] add German translations
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | no
| Deprecations? | no
| Tickets |
| License | MIT
| Doc PR |
Commits
-------
32d9a5298e add German translations
This PR was submitted for the master branch but it was merged into the 3.4 branch instead (closes#36192).
Discussion
----------
[Validator] Add french "at least" constraint translations
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | no
| Deprecations? | no
| License | MIT
Completion of PR https://github.com/symfony/symfony/pull/36165 by adding french translation for "at least" constraint.
Commits
-------
f885822350 Add french "at least" constraint translations
* 3.4:
[Http Foundation] Fix clear cookie samesite
[Security] Check if firewall is stateless before checking for session/previous session
[Form] Support customized intl php.ini settings
[Security] Remember me: allow to set the samesite cookie flag
[Debug] fix for PHP 7.3.16+/7.4.4+
[Validator] Backport translations
Prevent warning in proc_open()
This PR was merged into the 3.4 branch.
Discussion
----------
[Security/Http] Remember me: allow to set the samesite cookie flag
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
Similar to #35605, since Chrome 80 is going to require the `samesite` attribute.
This is a cherry-pick of #27976
Commits
-------
f0ceb73397 [Security] Remember me: allow to set the samesite cookie flag
This PR was squashed before being merged into the 3.4 branch.
Discussion
----------
[Http Foundation] Fix clear cookie samesite
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#36107
| License | MIT
With Chrome Update 80, Cookies are required to be `secure` and `samesite=none` for cross site requests. However they are defaulted to `samesite=lax` if the samesite attribute is not set. In other words: developer has to explicitely opt-in for `samesite=none` in the case of a cross site request.
More details: https://chromestatus.com/feature/5088147346030592
We add the `samesite` argument to `clearCookie` method to allow developer to explicitely set this value.
Commits
-------
4bdea1f2e7 [Http Foundation] Fix clear cookie samesite
This PR was submitted for the 4.4 branch but it was squashed and merged into the 3.4 branch instead.
Discussion
----------
[Security] Check if firewall is stateless before checking for session/previous session
| Q | A
| ------------- | ---
| Branch? | 4.4 <!-- see below -->
| Bug fix? | yes
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tickets | - <!-- prefix each issue number with "Fix #", if any -->
| License | MIT
| Doc PR | -
For one of our applications we had the issue that the session was always initialized, even for routes behind stateless firewalls. Using the redis session adapter this sometimes lead to exceptions if the connection failed. This change prevents the session from being initialized in the guard authentication handler for stateless firewalls
Commits
-------
9bb1230525 [Security] Check if firewall is stateless before checking for session/previous session
This PR was merged into the 3.4 branch.
Discussion
----------
[Form] Support customized intl php.ini settings
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | no
| License | MIT
| Doc PR | no
`IntlDateParser->parse()` behaves differently when `intl.error_level` and/or `intl.use_exceptions` are not 0.
This change makes sure `\IntlException` is caught when `intl.use_exceptions` is 1 and warnings thrown when `intl.error_level` is not 0 are ignored.
Commits
-------
61025d1d1b [Form] Support customized intl php.ini settings
`IntlDateParser->parse()` behaves differently when `intl.error_level` and/or `intl.use_exceptions` are not 0.
This change makes sure `\IntlException` is caught when `intl.use_exceptions` is 1 and warnings thrown when `intl.error_level` is not 0 are ignored.
This PR was merged into the 3.4 branch.
Discussion
----------
[Debug] fix for PHP 7.3.16+/7.4.4+
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
Should make tests green.
Follows https://github.com/php/php-src/pull/5206, which is part of PHP 7.4.4 and 7.3.16.
The same patch is needed on the ErrorHandler component in 4.4 of course.
Commits
-------
b3d9a8ac30 [Debug] fix for PHP 7.3.16+/7.4.4+
This PR was merged into the 4.4 branch.
Discussion
----------
[Mailer] Use %d instead of %s for error code in error messages
| Q | A
| ------------- | ---
| Branch? | 4.4 <!-- see below -->
| Bug fix? | no
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tickets | n/a <!-- prefix each issue number with "Fix #", if any -->
| License | MIT
| Doc PR | n/a
Commits
-------
9c3951ed69 [Mailer] Use %d instead of %s for error code in error messages
This PR was merged into the 4.4 branch.
Discussion
----------
[Security] Fixed hardcoded value of SODIUM_CRYPTO_PWHASH_MEMLIMIT_INTERACTIVE
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets |
| License | MIT
| Doc PR |
Commits
-------
677429479d [Security] Fixed hardcoded value of SODIUM_CRYPTO_PWHASH_MEMLIMIT_INTERACTIVE
This PR was squashed before being merged into the 3.4 branch.
Discussion
----------
Prevent warning in proc_open()
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | none
| License | MIT
| Doc PR | none
In addition to returning `false`, `proc_open()` triggers a warning when it fails. For example:
> Warning: proc_open(): fork failed - Cannot allocate memory
When using the `ErrorHandler`, the warning gets promoted to an exception, and the next line, `if (! is_resource(...`, is not executed. This mutes the warning and ensures that the next line is always executed and the proper exception is thrown.
Commits
-------
d43833a821 Prevent warning in proc_open()
This PR was squashed before being merged into the 4.4 branch.
Discussion
----------
[FrameworkBundle] Fix Router Cache
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#35932
| License | MIT
RouteCollection config cache didn't have the container file in the tracked resources. If container was recompiled, routes cache was not regenerated. This PR adds the container file to the route collection resources.
Commits
-------
c6ace13e34 [FrameworkBundle] Fix Router Cache
This PR was merged into the 4.4 branch.
Discussion
----------
[FrameworkBundle] Fix deprecation message for booting a kernel twice
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | no
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tickets | - <!-- prefix each issue number with "Fix #", 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/releases):
- Always add tests and ensure they pass.
- Never break backward compatibility (see https://symfony.com/bc).
- 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 master.
-->
Commits
-------
a0a6243a21 Fix deprecation messages
This PR was merged into the 4.4 branch.
Discussion
----------
[DI] fix preloading script generation
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
(fabbot failure is a false positive)
On master, we should work on being able to preload more classes (esp. all cache-warmup artifacts).
But for 4.4, this is good enough. Submitted as a bug fix because 1. the current code that deals with preloading kinda-works, but only on "dev" mode... and 2. fixing it provides a nice boost!
Small bench on a hello world:
- before: 380 req/s
- after: 580 req/s
That's +50%!
Pro-tip: adding a few `class_exists()` as done in this PR for the classes that are always used in the implementations (e.g. `new Foo()` in the constructor) will help the preload-script generator to work optimally. Without them, it will discover the symbols to preload only if they're found on methods.
Some of those `class_exists()` are mandatory, in relation to anonymous classes and https://bugs.php.net/79349
Commits
-------
a10fc4da5d [DI] fix preloading script generation
This PR was merged into the 4.4 branch.
Discussion
----------
[Security/Http] don't require the session to be started when tracking its id
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
`$session->getId()` returns the empty string when the session is not yet started.
When this happens, the session tracking logic wrongly detects that a new session was created and thus disables HTTP caching.
This fixes the issue by looking at the value of the session cookie instead.
(the case for `true` is when using `MockArraySessionStorage` as done in tests)
Commits
-------
c39188a7cc [Security/Http] don't require the session to be started when tracking its id
This PR was squashed before being merged into the 4.4 branch.
Discussion
----------
[DI] Fix CheckTypeDeclarationPass
| Q | A
| ------------- | ---
| Branch? | 4.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#35863 and #35972
| License | MIT
| Doc PR |
Bug 1: The lint container threw an error if a class buit with a factory was declared as callable while this factory method returne a callabe (#35863)
Bug 2: Sodium Exception was not caught in the CheckTypeDeclarationsPass. We have extended the exception caught to \Exception, instead of EnvNotFoundException and RuntimeException only.
Commits
-------
cbf4dfd10f [DI] Fix CheckTypeDeclarationPass
This PR was merged into the 3.4 branch.
Discussion
----------
[VarDumper] fix side-effect by not using mt_rand()
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
Reported by @bobthecow on Twitter: using `mt_rand()` breaks inspecting seeded calls to `mt_rand()`.
Should be replaced by a call to `md5(random_bytes(6))` on 4.4.
Commits
-------
8c85f91b9c [VarDumper] fix side-effect by not using mt_rand()