…t (julienfalque)

This PR was merged into the 5.4 branch.

Discussion
----------

[Finder] Add .gitignore nested negated patterns support

| Q             | A
| ------------- | ---
| Branch?       | 5.4
| Bug fix?      | no
| New feature?  | no
| Deprecations? | no
| Tickets       | -
| License       | MIT
| Doc PR        | -

Extracted from #43239.

Commits
-------

559639b [Finder] Add .gitignore nested negated patterns support

… (julienfalque)

This PR was merged into the 4.4 branch.

Discussion
----------

[Process] Run `open_basedir` tests in separate processes

| Q             | A
| ------------- | ---
| Branch?       | 4.4
| Bug fix?      | no
| New feature?  | no
| Deprecations? | no
| Tickets       | -
| License       | MIT
| Doc PR        | -

While working on #43239, I noticed that when a test sets `open_basedir`, the setting is not reset to its initial value after the test, which can have an impact on subsequent tests. I could not find the reason why the setting cannot be reset. I assume this is for security reasons, so the solution is to always run such test in a separate process.

This only concerns two tests for now. Before running those in separate process, the first one prevented subsequent tests in the same class from being run because they are skipped when `open_basedir` is set, which was always true.

Commits
-------

07fbfbf Run `open_basedir` tests in separate processes

…ectory (nickvergessen)

This PR was merged into the 6.4 branch.

Discussion
----------

[Finder] Also consider .git inside the basedir of in() directory

| Q             | A
| ------------- | ---
| Branch?       |6.4
| Bug fix?      | yes
| New feature?  | no
| Deprecations? | no
| Issues        | Fix described below
| License       | MIT

### Bug description
Previously when the `in()` directory was the root of your repository the `.gitignore` was skipped from consideration, which meant potentially all local files are ignored.

E.g. in the Nextcloud community we have the server locally with an `.gitignore` for the `apps/` directory. Now when an app developer develops their own app locally, they put their own git repository inside this ignored `apps/` folder.
The CS Fixer rules we use since a long time are:
```php
(new Finder())
	->ignoreVCSIgnored(true)
	->notPath('l10n')
	->notPath('vendor')
	->notPath('vendor-bin')
	->in(__DIR__);
```

and this worked pretty well with 5.4 and previous versions of the finder. Now that the Nextcloud project finally dropped PHP 8.0 for the latest version and we get the update to Symfony 6.4 the bug become visible locally, but our CI still works as we don't need the "server" repository as a parent when running the CS fixer.

### Locally - nested inside "server" repository
```
~/Nextcloud/server/apps/spreed$ composer run cs:check
> php-cs-fixer fix --dry-run --diff
PHP CS Fixer 3.54.0 15 Keys Accelerate by Fabien Potencier, Dariusz Ruminski and contributors.
PHP runtime: 8.2.18
Loaded config default from "/home/nickv/Nextcloud/30/server/appsbabies/spreed/.php-cs-fixer.dist.php".
Using cache file ".php-cs-fixer.cache".
    0 [░░░░░░░░░░░░░░░░░░░░░░░░░░░░]
```

:x:  No files found, all ignored by `server/.gitignore`

### CI without a parent repository
```
Run composer run cs:check || ( echo 'Please run `composer run cs:fix` to format your code' && exit 1 )
 > php-cs-fixer fix --dry-run --diff
PHP CS Fixer 3.54.0 15 Keys Accelerate by Fabien Potencier, Dariusz Ruminski and contributors.
PHP runtime: 8.1.2-1ubuntu2.13
Loaded config default from "/home/runner/actions-runner/_work/spreed/spreed/.php-cs-fixer.dist.php".
   0/502 [░░░░░░░░░░░░░░░░░░░░░░░░░░░░]   0%
```

:white_check_mark: Correct file list found

### 5.4
- From my POV this is a regression from #43239 (so 6.1+ only) and it becomes more "obvious" when backporting the unit test to 5.4 as it passes there without patching `VcsIgnoredFilterIterator.php`.
- Therefore cc-ing `@julienfalque` as author, to clarify if this was intentional

Commits
-------

af870c6 [Finder] Also consider .git inside the basedir of in() directory