Why add this test for 6.1? It is just a copy of one of the CommandDescriptionToPropertyRector tests

Okay so i added CommandPropertyToAttributeRector should i add a fixture of that test or should this be skipped as this is tested elsewhere?

If you ask me, the expected refactor for this deprecation should be a static property to an attribute.

Using setName() and setDescription() is not deprecated. The Symfony documentation just prefers it the "new" way for optimizations:

Defining the $defaultDescription static property instead of using the setDescription() method allows to get the command description without instantiating its class. This makes the php bin/console list command run much faster.

https://symfony.com/doc/current/console.html#configuring-the-command

So my personal expectation for using this set would be a refactors like this:

-#[\Symfony\Component\Console\Attribute\AsCommand(name: 'some:command']
+#[\Symfony\Component\Console\Attribute\AsCommand(name: 'some:command', description: 'Description')]
class SetDescriptionAtTop extends \Symfony\Component\Console\Command\Command
{
-    protected static $defaultDescription = 'Description';
}

and

+#[\Symfony\Component\Console\Attribute\AsCommand(name: 'some:command', description: 'Description')]
class SetDescriptionAtTop extends \Symfony\Component\Console\Command\Command
{
-    protected static $defaultName = 'some:command';
-    protected static $defaultDescription = 'Description';
}

But that last example has already been covered by the tests for CommandPropertyToAttributeRector, however we want to test changes using SymfonySetList::SYMFONY_61.

Maybe @TomasVotruba can guide us here. Do we need duplicate tests when a Rector is used in a set to verify behaviour of the set?

Maybe @TomasVotruba can guide us here. Do we need duplicate tests when a Rector is used in a set to verify behaviour of the set?

That's not necessary indeed. But the bug should be fixed, so CI pass.

Check i added missing stubs