This PR was merged into the 4.2-dev branch.
Discussion
----------
[Cache] improve perf when using RedisCluster by reducing roundtrips to the servers
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
Improves perf when using RedisCluster by:
- disabling versioning and replacing by per-master clear/flush
- grouping multiple "get" in one "mget"
- enabling pipelining in PredisCluster mode
~Might need adjustment depending on the answer to https://github.com/nrk/predis/issues/520~
Commits
-------
5155f48029 [Cache] improve perf when using RedisCluster by reducing roundtrips to the servers
This PR was merged into the 4.2-dev branch.
Discussion
----------
[Serializer] Add support for ignoring comments while XML encoding
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | /
| License | MIT
| Doc PR | /
In addition to https://github.com/symfony/symfony/pull/27926 which allowed to ignore XML processing instructions, this PR allows to ignore the XML comments while encoding.
Commits
-------
8f8230ac3a Add support for ignoring comments while XML encoding
This PR was merged into the 4.2-dev branch.
Discussion
----------
[Messenger] Remove the "obscure" message subscriber configuration
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | no
| BC breaks? | yes
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | ø
| License | MIT
| Doc PR | ø
As described in #28275, all of the configuration can be done using yield and that we could remove the support for other ways (especially the obscure return `[['method', -10]]` syntax) as I believe this would clarify the configuration a lot.
Commits
-------
cf2ad861f5 Remove the "obscure" message subscriber configuration
This PR was merged into the 4.2-dev branch.
Discussion
----------
[Messenger] Allow interfaces to be type-hinted as well
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #27076
| License | MIT
| Doc PR | ø
Interfaces can be type-hinted as well for the message handlers.
Commits
-------
2dbbfbda4e Allow interfaces to be type-hinted as well
This PR was merged into the 4.2-dev branch.
Discussion
----------
[Messenger] Add a --bus option to the messenger:consume-messages command
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | n/a
| License | MIT
| Doc PR | todo
Making it compatible with the multi-bus feature.
Commits
-------
e3f1eecbc1 Bus argument is a required option when multiple buses are defined
539cb62ffe [Messenger] Add a --bus option to the messenger:consume-messages command
This PR was merged into the 4.2-dev branch.
Discussion
----------
[Messenger] Only subscribe to a given bus from the MessageSubscriber
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #... <!-- #-prefixed issue number(s), if any -->
| License | MIT
| Doc PR | ø
#27275 introduced the ability to listen to only a few buses from the handler tag. This adds that ability directly from the message subscriber.
It has also highlighted to me that most of the configuration can be done using `yield` (like the example I've added in this PR's tests) and that we could remove the support for other ways (especially the obscure `return [['method', -10]]` syntax) but I believe this should be done **in another pull-request** (that I'm happy to do after this one).
Commits
-------
f60e409011 Only subscribe to a given bus from the MessageSubscriber
This PR was squashed before being merged into the 4.2-dev branch (closes#28070).
Discussion
----------
[Translator] Use ICU parent locales as fallback locales
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | #12319
| License | MIT
| Doc PR | https://github.com/symfony/symfony-docs/pull/10122
Currently the `Translator` fall backs based on the locale separator (eg `es_AR` to `es`), but the ICU data contains parent locales (eg `es_AR` is a child of `es_419`, as is `es_BO`, `es_EC` etc).
This makes use of the ICU data to add add in these fallbacks. This means the specific locales can be used, but the translations can stored in these groupings (eg `es_419` for Latin American Spanish), as well as adding other sensible fallbacks (eg Cape Verdean Portuguese to `pt_PT`).
Commits
-------
e0f402fc29 [Translator] Use ICU parent locales as fallback locales
* 4.1:
bumped Symfony version to 2.8.46
updated VERSION for 2.8.45
update CONTRIBUTORS for 2.8.45
updated CHANGELOG for 2.8.45
[PhpUnitBridge] keep compat with composer 1.0
Instantiate $offset and $maxlen at definition
[Cache] minor code update to leverage PHP 7.1
* 3.4:
bumped Symfony version to 2.8.46
updated VERSION for 2.8.45
update CONTRIBUTORS for 2.8.45
updated CHANGELOG for 2.8.45
[PhpUnitBridge] keep compat with composer 1.0
Instantiate $offset and $maxlen at definition
* 2.8:
bumped Symfony version to 2.8.46
updated VERSION for 2.8.45
update CONTRIBUTORS for 2.8.45
updated CHANGELOG for 2.8.45
Instantiate $offset and $maxlen at definition
This PR was merged into the 4.2-dev branch.
Discussion
----------
[VarExporter] a new component to serialize values to plain PHP code
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
This PR proposes moving what is currently the `PhpMarshaller` class in the Cache component to a separate component.
This component would provide only one public static method:
`VarExporter::export($value, bool &$isStaticValue = null): string`.
This method returns `$value` serialized as plain PHP code. Running this code creates the same exact data structure that `$value` contained. This is exactly like `serialize()` and `unserialize()`, from which all semantics are preserved (`__sleep`, `__wakeup` and `Serializable`).
The reason to use this method *vs* `serialize()` or even igbinary is performance: thanks to OPcache, the resulting code is significantly faster and more memory efficient than using `unserialize()` or `igbinary_unserialize()`.
Unlike `var_export()`, this works on any serializable PHP value.
It also provides a few improvements over `var_export()`/`serialize()`:
- the output is PSR-2 compatible
- the output can be re-indented without messing up with any `\r` or `\n` in the data
- missing classes throw a `ReflectionException` instead of being unserialized to a `PHP_Incomplete_Class` object
- references involving `SplObjectStorage`, `ArrayObject` or `ArrayIterator` instances are preserved
- `Reflection*`, `IteratorIterator` and `RecursiveIteratorIterator` classes throw an exception when being serialized (their unserialized version is broken anyway, see https://bugs.php.net/76737.)
Commits
-------
7831ad75e5 [VarExporter] a new component to serialize values to plain PHP code
This PR was merged into the 4.2-dev branch.
Discussion
----------
[FrameworkBundle] Added new "auto" mode for `framework.session.cookie_secure` to turn it on when https is used
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
I'm pretty sure we're many forgetting to make session cookies "secure".
Here is an "auto" mode that makes them secure automatically when the session is started on requests with the "https" scheme.
Commits
-------
4f7b41a83d [FrameworkBundle] Added new "auto" mode for `framework.session.cookie_secure` to turn it on when https is used
This PR was merged into the 2.8 branch.
Discussion
----------
[HttpFoundation] Fix unprepared BinaryFileResponse sends empty file
| Q | A
| ------------- | ---
| Branch? | 2.8
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes, with the exception of preexisting, unrelated failures
| Fixed tickets | #28237
| License | MIT
| Doc PR |
When you call `BinaryFileResponse#sendContent()` without first calling `prepare()` the response is sent but the contents are empty. `prepare()` properly initializes the `$maxlen` and `$offset` properties. However, `sendContent()` doesn't do any sanity checking, and so, uses the uninitialized properties. This causes `stream_copy_to_stream()` to copy empty contents and the file that is sent, to contain nothing.
This change initializes the properties at definition instead of in `prepare()`.
> Additionally:
> - Bug fixes must be submitted against the lowest branch where they apply
~I'm not sure how early this bug exists, or how far back to go. I'll check to see if 2.7 and 2.8 are affected and report back.~
Commits
-------
dba8687a5d Instantiate $offset and $maxlen at definition
This PR was merged into the 4.2-dev branch.
Discussion
----------
[Serializer] Update changelog about the new $encoderIgnoredNodeTypes arg in XmlEncoder contrustor
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | https://github.com/symfony/symfony/pull/27926
| License | MIT
| Doc PR | /
Commits
-------
49f3bfc823 Update changelog
This PR was merged into the 4.2-dev branch.
Discussion
----------
[Messenger] Don't make EnvelopeItemInterface extend Serializable
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | no
| BC breaks? | yes (on experimental API)
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | -
| License | MIT
| Doc PR | -
`Serializable` is a broken interface, see e.g. https://externals.io/message/98834
I don't think we should force ppl to implement it anywhere.
Actually, it isn't required to be able to serialize an object, and it doesn't enforce making a class serializable (as the `serialize()` method can throw).
What was the purpose of the removed logic? ping @sroze @ogizanagi
Commits
-------
2beda894f2 [Messenger] Don't make EnvelopeItemInterface extend Serializable
This PR was merged into the 4.2-dev branch.
Discussion
----------
Mock date() in ClockMock
| Q | A
| ------------- | ---
| Branch? | master
| Bug fix? | no
| New feature? | yes
| BC breaks? | no (In case `date()` is mocked in some other way execution would fail because of redeclaration. Could be avoided with an extra `function_exists()` check. WDYT?)
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets |
| License | MIT
| Doc PR |
As to the [PHP documentation](https://secure.php.net/manual/en/function.date.php) `date()` uses the value of `time()` as timestamp if none is given. `date()` has to be mocked in ClockMock as well for this still being true, otherwise `\time()` is used as default.
BTW: The msec part of `microtime()` has 8 fractional digits on my system, ClockMock returns only 6...
Commits
-------
e8ba79adb9 [Bridge/PhpUnit] Mock date() in ClockMock
* 4.1:
[DoctrineBridge] support __toString as documented for UniqueEntityValidator
[travis] enable Redis cluster
[Cache] enable Memcached::OPT_TCP_NODELAY to fix perf of misses
fix data mapper return type in docblock
fix type error handling when writing values
* 3.4:
[DoctrineBridge] support __toString as documented for UniqueEntityValidator
[travis] enable Redis cluster
[Cache] enable Memcached::OPT_TCP_NODELAY to fix perf of misses
fix data mapper return type in docblock
fix type error handling when writing values