This PR was merged into the 3.4 branch.
Discussion
----------
[DI] Improve performance of processDefinition
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | kind of
| New feature? | no
| Deprecations? | no
| Tickets | no
| License | MIT
| Doc PR |
Saves some time during container compilation by instantiating the reflection class only once. In my case this speeds up container compilation in dev mode by ~10% (saves almost 100k calls to `getReflectionClass`).
Tests still run locally and my compiled container was identical pre and post change, but I found this improvement by Blackfire profiling and am not familiar with the surrounding code, so it would be great if someone could doublecheck if the change causes problems.
Commits
-------
41b56eac29 [DI] Improve performance of processDefinition
This PR was merged into the 3.4 branch.
Discussion
----------
Fix regex lookahead syntax in ApplicationTest
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | no (test fix)
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
`(?>!1)` is actually an atomic group for literal `!1`.
`(?!1)` is a negative lookahead assertion for literal `1`.
Commits
-------
cf1d7b2ed1 Fix regex lookahead syntax in ApplicationTest
This PR was merged into the 3.4 branch.
Discussion
----------
[SecurityBundle][FirewallMap] Remove unused property
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | no
| New feature? | no
| Deprecations? | no
| Tickets | no
| License | MIT
| Doc PR | -
This property is unused. It was removed in https://github.com/symfony/symfony/pull/22943. It was likely reintroduced in a merge commit. It is still on master.
Commits
-------
0904e57b85 [SecurityBundle][FirewallMap] Remove unused property
This PR was merged into the 3.4 branch.
Discussion
----------
[Validator][ConstraintValidator] Safe fail on invalid timezones
Co-authored-by: Scott Dawson <scott@loyaltycorp.com.au>
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | https://github.com/symfony/symfony/issues/33901
| License | MIT
| Doc PR |
Alternative to https://github.com/symfony/symfony/pull/33902.
I will explain why I think it is better this way:
1. We set the timezone with the setter because it's 100% safe, it never fails. It fall backs to the default timezone if the provided timezone is not supported (as if we passed null, so the same behavior that always existed). We are therefore compatible with all edge cases.
2. We don't validate the timezone with `\DateTimeZone::listIdentifiers()`. It only returns full identifiers like "Europe/Paris" but it doesn't take into account "numeric" identifiers such as "+08:00" which are perfectly valid. I added a test case to ensure we stay valid with this case. + some invalid identifiers for the native `\IntlDateFormatter` are valid with the polyfill that uses `\DateTimeZone` (eg : `X`). I don't think we can validate anything safely that will work reliably on both implementations.
Commits
-------
3b1b994cb3 [Validator][ConstraintValidator] Safe fail on invalid timezones
This PR was merged into the 3.4 branch.
Discussion
----------
[DoctrineBridge] try to fix deprecations from doctrine/persistence
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
Follows https://github.com/doctrine/persistence/pull/71
But the BC layer is not working yet, as highlighted by the `XXX` in the attached patch.
At least for the corresponding interfaces, doctrine/persistence should always alias the legacy name to the new one.
/cc @greg0ire @alcaeus FYI
Commits
-------
53a4711520 [DoctrineBridge] try to fix deprecations from doctrine/persistence
This PR was merged into the 3.4 branch.
Discussion
----------
[Translation] fix memoryleak in PhpFileLoader
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
This happens when running the test suite with opcache disabled (as it is the case by default since `opcache.enable_cli=0`).
Doing this "require" in a loop (for each test case) compiles the file at each iteration and doesn't reclaim memory (there is no garbage collector for opcodes).
Commits
-------
5c9e3bac96 [Translation] fix memoryleak in PhpFileLoader
This PR was merged into the 3.4 branch.
Discussion
----------
[HttpFoundation] Use `Cache-Control: must-revalidate` only if explicit lifetime has been given
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets |
| License | MIT
| Doc PR |
This is really nit-picking: The conservative, safe default for `Cache-Control` is `private, no-cache` which means the response must not be served from cache unless it has been validated.
If `Last-Modified` or `Expires` are present, we can relax `no-cache` to be `must-revalidate`, which means that _once the response has become stale_, it must be revalidated.
An `ETag` alone does not give the response a lifetime, so IMO sticking with `no-cache` in this case would be more consistent.
Commits
-------
1b1002b426 [HttpFoundation] Use `Cache-Control: must-revalidate` only if explicit lifetime has been given
This PR was squashed before being merged into the 3.4 branch.
Discussion
----------
[Yaml] Implement multiline string as scalar block for tagged values
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| Tickets | Fix #... <!-- prefix each issue number with "Fix #", if any -->
| License | MIT
At the moment you can parse a tagged value defined as a scalar block. But you can't actually dump a multiline string as scalar block when using a tagged value.
This PR implements the multiline string as scalar block for tagged values as well.
Commits
-------
84241d4e62 [Yaml] Implement multiline string as scalar block for tagged values
This PR was merged into the 3.4 branch.
Discussion
----------
[MonologBridge] Fix debug processor datetime type
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets |n/a
| License | MIT
| Doc PR | n/a
This PR fixes an eventual `Call to a member function getTimestamp() on string` if any Processor transforms the `datetime` value, and uses the same record conditions as in the [ConsoleFormatter](https://github.com/symfony/monolog-bridge/blob/master/Formatter/ConsoleFormatter.php#L118 )
Commits
-------
ffe3f10780 [MonologBridge] Fix debug processor datetime type
This PR was merged into the 3.4 branch.
Discussion
----------
[CI] Replace php7.4snapshot with php7.4 in Travis configuration
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | no
| New feature? | no
| Deprecations? | no
| Tickets | no
| License | MIT
| Doc PR | no
There is a php7.4 already supported on Travis CI. This pull request replaces previous `7.4snapshot` with `7.4` in Travis config.
Commits
-------
4d3b73e688 [CI] Replace php7.4snapshot with php7.4 in Travis configuration
This PR was squashed before being merged into the 3.4 branch.
Discussion
----------
[FrameworkBundle] Use UserInterface to @return in getUser method
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | no
| New feature? | no <!-- please update src/**/CHANGELOG.md files -->
| Deprecations? | no <!-- please update UPGRADE-*.md and src/**/CHANGELOG.md files -->
| License | MIT
<!--
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):
- 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
-------
9f1ebd7fb9 [FrameworkBundle] Use UserInterface to @return in getUser method
This PR was merged into the 3.4 branch.
Discussion
----------
[ExpressionLanguage] Process division by zero
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets |
| License | MIT
| Doc PR |
To be able to catch the error in expression like ` 1 / 0`
**Before PR:**
```
try {
1 / 0;
} catch (\Throwable $e) {
// It won't be caught anyway
// PHP Warning: Division by zero in...
}
try {
1 % 0;
} catch (\Throwable $e) {
// It will be caught since PHP7
// \DivisionByZeroError with message `Modulo by zero`
}
```
**After PR:**
```
try {
1 / 0;
} catch (\Throwable $e) {
// It will be caught
// \DivisionByZeroError with message `Division by zero`
}
try {
1 % 0;
} catch (\Throwable $e) {
// It will be caught
// \DivisionByZeroError with message `Modulo by zero`
}
```
Commits
-------
02ab72ab30 [ExpressionLanguage][Node][BinaryNode] Process division by zero
This PR was merged into the 3.4 branch.
Discussion
----------
[Validator][ConstraintValidator] Stop passing unnecessary timezone argument to \DateTime
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | no
| New feature? | no
| Deprecations? | no
| Tickets | -
| License | MIT
| Doc PR | -
Since we format the \DateTimeImmutable with the "e" character, it uses this timezone identifier and do not consider the passed one. See https://www.php.net/manual/en/datetime.construct.php:
> The $timezone parameter and the current timezone are ignored when the $time parameter either is a UNIX timestamp (e.g. @946684800) or specifies a timezone (e.g. 2010-01-28T15:00:00+02:00).
Commits
-------
03dbcf8794 [Validator][ConstraintValidator] Stop passing unnecessary timezone argument to \DateTime
This PR was merged into the 3.4 branch.
Discussion
----------
[TwigBundle] add tags before processing them
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#34815
| License | MIT
| Doc PR |
Commits
-------
e1145a78b5 add tags before processing them
This PR was merged into the 3.4 branch.
Discussion
----------
[Cache] fix memory leak when using PhpArrayAdapter
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#34687
| License | MIT
| Doc PR | -
Thanks to @adrienfr, I've been able to understand what causes this massive memory leak when using `PhpArrayAdapter`:
When tests run, a new kernel is booted for each test case. This means a new instance of `PhpArrayAdapter` is created, which means it loads its state again and again using `include` for e.g. `annotations.php` in this example.
The first obvious thing is that we see this doing `compile::*`: this means PHP is parsing the same file again and again. But shouldn't opcache prevent this? Well, it's disabled by default because `opcache.enable_cli=0`. To prove the point, here is a comparison with the same tests run with `php -dopcache.enable_cli=1`. The comparison is swapped, but you'll get it:
But that's not over: because of https://bugs.php.net/76982 (see #32236 also), we still have a memory leak when the included file contains closures. And this one does.
This PR fixes the issue by storing the return value of the include statement into a static property. This fits the caching model of `PhpArrayAdapter`: it's a read-only storage for system caches - i.e. its content is immutable.
Commits
-------
4194c4c56d [Cache] fix memory leak when using PhpArrayAdapter
This PR was merged into the 3.4 branch.
Discussion
----------
[SecurityBundle] Properly escape regex in AddSessionDomainConstraintPass
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | https://github.com/symfony/symfony/issues/34774
| License | MIT
| Doc PR | -
`%s` should be escaped, so it is dumped as `%%s` (it ends up being properly unescaped at load time, so the passed value to the service is the same).
Commits
-------
de03cee846 [SecurityBundle] Properly escape regex in AddSessionDomainConstraintPass
This PR was merged into the 3.4 branch.
Discussion
----------
[FrameworkBundle] resolve service locators in `debug:*` commands
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#34470
| License | MIT
| Doc PR | -
Because of the way ServiceClosureArgument are dumped, we need to resolve locators after loading the xml dump of the container:
https://github.com/symfony/symfony/blob/3.4/src/Symfony/Component/DependencyInjection/Dumper/XmlDumper.php#L273
Commits
-------
820da66346 [FrameworkBundle] resolve service locators in `debug:*` commands
This PR was merged into the 3.4 branch.
Discussion
----------
[3.4][Validator] Allow underscore character "_" in URL username and password
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| License | MIT
Hello!
It's been a long time since my last push on Symfony :)
Here's a bug fix. I think URL usernames and password may contain an underscore. Let me know!
Commits
-------
869518bc7e [Validator] Allow underscore character "_" in URL username and password
This PR was merged into the 3.4 branch.
Discussion
----------
[SecurityBundle] Passwords are not encoded when algorithm set to "true"
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#34725
| License | MIT
| Doc PR | -
If the algorithm is set to `true`, password will be encode as plain password.
```
security:
encoders:
App\User\User:
algorithm: true
```
The reason for this is the not strict comparison of php switches.
```
switch ($config['algorithm']) {
case 'plaintext':
}
```
`true == 'plaintext'` is `true`, so the first case is hit. My first solution was to cast the algorithm to a string, to prevent this. After some feedback I have catch this problem earlier and does not allow true as valid value to the algorithm option.
Ps. This is my first PR for Symfony, any feedback is welcome :-)!
Commits
-------
83a5517c01 [SecurityBundle] Passwords are not encoded when algorithm set to \"true\"
This PR was merged into the 3.4 branch.
Discussion
----------
[Security] do not validate passwords when the hash is null
| Q | A
| ------------- | ---
| Branch? | 3.4
| Bug fix? | yes
| New feature? | no
| Deprecations? | no
| Tickets | Fix#34775
| License | MIT
| Doc PR |
Commits
-------
5699cb22bb do not validate passwords when the hash is null
This PR was merged into the 3.4 branch.
Discussion
----------
Allow copy instead of symlink for ./link script
| Q | A
| ------------- | ---
| Branch? | 3.4 <!-- 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
Not the most efficient way to work, but sometimes it helps to test a bug fix/feature within an existing project for which symlinks can't be resolved due to the dev environment (e.g: a Vagrant where only the current project directory is mounted).
Commits
-------
b28fe66363 Allow copy instead of symlink for ./link script
