…sible (weaverryan)

This PR was merged into the 6.3 branch.

Discussion
----------

[AssetMapper] Improving XSD to use attributes whenever possible

| Q             | A
| ------------- | ---
| Branch?       | 6.3
| Bug fix?      | yes
| New feature?  | no
| Deprecations? | no
| Tickets       | None
| License       | MIT
| Doc PR        | TODO

Some notes from `@stof` - I was using `<elements>` for scalar config that works much better as attributes.

Cheers!

Commits
-------

6326a5e [AssetMapper] Improving XSD to use attributes whenever possible

This PR was squashed before being merged into the 6.3 branch.

Discussion
----------

[AssetMapper] Add cached asset factory

| Q             | A
| ------------- | ---
| Branch?       | 6.3
| Bug fix?      | no
| New feature?  | yes
| Deprecations? | no
| Tickets       | None
| License       | MIT
| Doc PR        | Still TODO

Hi!

This is built on top of #50264. I CAN separate them if needed - sorry for the noise.

tl;dr When using asset mapper, to load the page, we need to build the importmap & resolve a few paths, like `{{ asset('styles/app.css') }}`. Because asset mapper loads the contents and performs some regex on CSS and JS files, this can slow down the page in dev (in prod, `asset-map:compile` removes all overhead).

This PR fixes that (I can see a big improvement in my test app). We "cache" the `MappedAsset` objects via the config cache system so that if an underlying file changes (e.g. `assets/styles/app.css`), we only need to rebuild that ONE `MappedAsset` when the page is loading.

Cheers!

Commits
-------

ce77ed8 [AssetMapper] Add cached asset factory

released v6.3.0-BETA3

* 5.4:
  [HttpKernel] Account for Response::getDate() possibly returning a DateTimeImmutable
  [Messenger] Respect `isRetryable` decision of the retry strategy when deciding if failed message should be re-delivered

* 6.2:
  [HttpKernel] Account for Response::getDate() possibly returning a DateTimeImmutable
  [Messenger] Respect `isRetryable` decision of the retry strategy when deciding if failed message should be re-delivered

This PR was merged into the 6.3 branch.

Discussion
----------

[Notifier] Bring consistency to bridges

| Q             | A
| ------------- | ---
| Branch?       | 6.3
| Bug fix?      | no
| New feature?  | no
| Deprecations? | no
| Tickets       | -
| License       | MIT
| Doc PR        | -

Commits
-------

657e9be [Notifier] Bring consistency to bridges

* 5.4:
  [FrameworkBundle] Generate caches consistently on successive run of `cache:clear` command

* 6.2:
  [FrameworkBundle] Generate caches consistently on successive run of `cache:clear` command

…epelko)

This PR was submitted for the 6.2 branch but it was merged into the 6.3 branch instead.

Discussion
----------

[Console] Remove redundant method getSaturation()

| Q             | A
| ------------- | ---
| Branch?       | 6.2
| Bug fix?      | yes
| New feature?  | no
| Deprecations? | no
| Tickets       | none
| License       | MIT
| Doc PR        | none

1. in `getSaturation()`, `$diff` is compared against `int` zero. This will only yield `true` when RGB values contain min/max color value (`0` or `255`). It will be always `false` for floats.
2. in `degradeHexColorToAnsi4()` value returned from `getSaturation()` is `round`ed, which returns float, thus making comparison to `int` zero useless.
3. in the same method, bitwise operations (final `return` statement) will convert number to `int` anyway, so the explicit cast was removed

In conclusion, `getSaturation` is redundant. Tests are expanded to show that even in min/max cases everything works fine.

Commits
-------

38021b5 [Console] Remove redundant method getSaturation()

This PR was merged into the 6.3 branch.

Discussion
----------

Ensure DoctrineIntegrationTest always run in UTC

| Q             | A
| ------------- | ---
| Branch?       | 6.3
| Bug fix?      | yes
| New feature?  | no
| Deprecations? | no
| Tickets       | Fix #50327
| License       | MIT
| Doc PR        | N/A

Commits
-------

c036acf Ensure DoctrineIntegrationTest always run in UTC

This PR was merged into the 6.3 branch.

Discussion
----------

[VarDumper] Fix HTML of invisible characters

| Q             | A
| ------------- | ---
| Branch?       | 6.3
| Bug fix?      | yes
| New feature?  | no
| Deprecations? | no
| Tickets       | -
| License       | MIT
| Doc PR        | -

Commits
-------

6b0ebc8 [VarDumper] Fix HTML of invisible characters

…value if possible when a parameter is not found (MatTheCat)

This PR was merged into the 6.3 branch.

Discussion
----------

[HttpKernel] Make `QueryParameterValueResolver` provide a value if possible when a parameter is not found

| Q             | A
| ------------- | ---
| Branch?       | 6.3
| Bug fix?      | yes
| New feature?  | no
| Deprecations? | no
| Tickets       | Fix #50339
| License | MIT
| Doc PR        | N/A

As `QueryParameterValueResolver` is “targeted” it must provide a value itself if possible.

When a parameter is not found in the query string, we want to return the argument’s default value if it has one, or `null` if it is nullable. Else the behavior do not change: we throw a `NotFoundHttpException`.

Commits
-------

00df479 Provide a value if possible when parameter is not found

…medJsonResponse (alexander-schranz)

This PR was merged into the 6.3 branch.

Discussion
----------

[HttpFoundation] Fix problem with empty generator in StreamedJsonResponse

| Q             | A
| ------------- | ---
| Branch?       | 6.3 (Feature `StreamedJsonResponse`: #47709)
| 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 - was reported to me on Slack by `@norkunas`
| License       | MIT
| Doc PR        | symfony/symfony-docs#... <!-- required for new features -->

Currently when the Generator is empty the return is invalid JSON which should not happen. So adding a testcase and a fix to the problem with the empty generator.

Commits
-------

39bb6b6 Fix problem with empty generator in StreamedJsonResponse

…view (javiereguiluz)

This PR was merged into the 6.3 branch.

Discussion
----------

[WebProfilerBundle] Tweak the HTML code of the Twig entry view

| Q             | A
| ------------- | ---
| Branch?       | 6.3
| Bug fix?      | yes
| New feature?  | no
| Deprecations? | no
| Tickets       | -
| License       | MIT
| Doc PR        | -

Stretched links are used to make entire sections, like a card, clickable. See https://getbootstrap.com/docs/5.3/helpers/stretched-link/

Maybe `@PhilETaylor` can answer us about this (he made the original PR to add this in #49887) but I think it wasn't intended to make entire Twig panel clickable to open the template. So, in this PR I propose to remove this behavior.

Here you can see how in the Security panel, the link is not stretched but in the Twig panel it is:

![profiler-panel](https://github.com/symfony/symfony/assets/73419/155170f3-115d-4043-98cd-8844c1a3a4bf)

Commits
-------

c9ed471 [WebProfilerBundle] Tweak the HTML code of the Twig entry view

… in `HtmlDumper` (ohader)

This PR was squashed before being merged into the 6.3 branch.

Discussion
----------

[VarDumper] Reduce stylesheet assignments via JavaScript in `HtmlDumper`

| Q             | A
| ------------- | ---
| Branch?       | 6.3
| Bug fix?      | no
| New feature?  | no
| Deprecations? | no
| Tickets       | *none*
| License       | MIT

The dynamic `<style>` element in the JavaScript variable `refStyle` can be replaced by static CSS and element states.
The new rule `.sf-dump-hover:hover` substitutes JavaScript event handling for `mouseover` events.

This is a preparation to have the possibility to assign `nonce` HTML attributes to inline `<script>` and `<style>` nodes, e.g. shown as proof-of-concept at
https://review.typo3.org/c/Packages/TYPO3.CMS/+/78512/2/typo3/sysext/adminpanel/Classes/Utility/HtmlDumper.php

Commits
-------

53046a3 [VarDumper] Reduce stylesheet assignments via JavaScript in `HtmlDumper`

This PR was merged into the 6.2 branch.

Discussion
----------

[Clock] Fix MockClock::modify() on PHP 8.3

| Q             | A
| ------------- | ---
| Branch?       | 6.2
| Bug fix?      | yes
| New feature?  | no
| Deprecations? | no
| Tickets       | -
| License       | MIT
| Doc PR        | -

Commits
-------

7361374 [Clock] Fix MockClock::modify() on PHP 8.3

* 6.2:
  [Clock] Fix MockClock::modify() on PHP 8.3

This PR was merged into the 6.3 branch.

Discussion
----------

[HttpClient] Remove final keyword on `AsyncResponse`

| Q             | A
| ------------- | ---
| Branch?       | 6.3
| Bug fix?      | kind of
| New feature?  | no
| Deprecations? | no
| Tickets       |
| License       | MIT
| Doc PR        |

In symfony < 6.3, it was possible to override HTTP response headers. See this [blog post](https://jolicode.com/blog/aggressive-caching-with-symfony-http-client) for a use case

With Symfony 6.3, this is not possible anymore. I created a reproducer here: https://github.com/lyrixx/test/tree/http-cache-pp

And `@nicolas`-grekas came with another solution: lyrixx/test#11

So here we go!

Commits
-------

fbd6d18 [HttpClient] Remove final keyword on AsyncResponse

…vars (Okhoshi)

This PR was merged into the 6.3 branch.

Discussion
----------

[DependencyInjection] Fix support for `false` boolean env vars

| Q             | A
| ------------- | ---
| Branch?       | 6.3
| Bug fix?      | yes
| New feature?  | no
| Deprecations? | no
| Tickets       | -
| License       | MIT
| Doc PR        | -

After upgrading `symfony/dependency-injection` package to version 6.3, some of our env vars couldn't be found anymore.

For some scripts, we are providing an adhoc default value to env vars by setting `$_SERVER` directly. However, since version 6.3 of the DependencyInjection component, setting an env var to `false` (the boolean value, like in the snippet below) doesn't work anymore, and Symfony reports that the variable cannot be found.

```php
$_SERVER['FOO'] = false;
```

It seems to be a side effect of the changes made in #48705.

Commits
-------

5101d18 [DependencyInjection] Fix support for `false` boolean env vars

…ltzu)

This PR was merged into the 6.3 branch.

Discussion
----------

[Messenger] Don't mark `RedispatchMessage` as internal

| Q             | A
| ------------- | ---
| Branch?       | 6.3
| Bug fix?      | yes
| New feature?  | no
| Deprecations? | no
| License       | MIT

As discussed in #49734 (comment), `RedispatchMessage` is meant to be used in your project, hence it shouldn't be marked as internal (which makes IDEs suggest not to use it).

Commits
-------

1209b81 Don't mark RedispatchMessage as internal

This PR was merged into the 6.3 branch.

Discussion
----------

[Security] Fix log message in `OidcTokenHandler`

| Q             | A
| ------------- | ---
| Branch?       | 6.3
| Bug fix?      | no
| New feature?  | no
| Deprecations? | no
| Tickets       | -
| License       | MIT
| Doc PR        | -

Commits
-------

5f29bd8 [Security] Fix log message in OidcTokenHandler

When using `$platform->getAlterSchemaSQL($schemaDiff)` instead of `$schemaDiff->toSaveSql($platform)` causes to call `\Doctrine\DBAL\Schema\SchemaDiff::_toSql($platform, false)` instead of `\Doctrine\DBAL\Schema\SchemaDiff::_toSql($platform, true)`.  When $saveMode=false the diff from the schema is getting remove, so actually all other tables ... in the DB are getting deleted