BUGFIX: Adjust to Php 83 get_parent_class deprecation

[mhsdesign]

see https://www.php.net/manual/en/function.get-parent-class.php

Upgrade instructions

Review instructions

Checklist

• Code follows the PSR-2 coding style
• Tests have been created, run and adjusted as needed
• The PR is created against the lowest maintained branch
• Reviewer - PR Title is brief but complete and starts with FEATURE|TASK|BUGFIX
• Reviewer - The first section explains the change briefly for change-logs
• Reviewer - Breaking Changes are marked with !!! and have upgrade-instructions

[mhsdesign]

Must have gone missing while upmerging #3032 i guess

[lorenzulrich]

@mhsdesign I suspect that this change has regressions. In #3406, I backported this change to 8.3 and applied it as a patch in an installation.

Now, when I call an own repository that extends a Neos repository and call a method that creates a query, I get this error:

Call to a member function createQueryForType() on null

  Type: Error
  File: Packages/Framework/Neos.Flow/Classes/Persistence/Repository.php
  Line: 137

So somehow DI fails in this case. I suspect it is because PHP <8.3 get_parent_class() is not identical to PHP 8.3 get_parent_class($this).

https://www.php.net/manual/en/function.get-parent-class.php#refsect1-function.get-parent-class-changelog

[lorenzulrich]

It seems that using method_exists(parent::class as mentioned here seems to work in my situation. However I'd be surprised if that part of Flow wasn't covered by a test.

[kdambekalns]

As mentioned in the post you linked, parent::class fails without a parent and the proper replacement would be get_parent_class($this). Also the PHP documentation says:

previously, calling this function inside a class returned the name of that class.

And inside a class, $this is, well, "that class". Also, https://3v4l.org/6TtkP shows no difference for PHP < 8.3

[kitsunet]

Given that this here happens in a proxy class we can safely assume a parent::class cannot fail?!

[kdambekalns]

Yep, in this case, true. I was more trying to prove that the change as a reason for the "mthod call on null" seems… strange?!