Review main Merged

91192[BUGFIX] Prevent collision of extbase parameters with level delimiter

Nikita Hovratov CR +2 V +2 Normal +140 −28 updated 04/08/26, 18:40

Commit message

Extbase's internal arguments already use the double underscore prefix.
This collides with the VariableProcessor LEVEL_DELIMITER, which is the
same and mistakes these arguments to be a part of the processed nesting
structure.

This patch adds a third underscore, to have a unique delimiter.

Review

  • Code review CR +2 Done
    • Christian Kuhn +2
    • Garvin Hicking +1
  • Verified V +2 Done
    • Christian Kuhn +2
    • Garvin Hicking +1
    • core-ci +1 CI

Issues 1

Files 3

Review log 1925

  1. Nikita Hovratov uploaded patch set 1 10/22/25, 08:12

  2. core-ci voted on patch set 1 V +1 10/22/25, 08:22 Core CI is happy: https://git.typo3.org/typo3/CI/cms/-/pipelines/87219

  3. Nikita Hovratov rebased onto patch set 2 V +1 12/09/25, 17:57

  4. Christian Kuhn commented on patch set 2 12/09/25, 18:03

    args. i wonder this did not pop up earlier. you're sure i triggered/caused this with the main extbase request patch?
  5. core-ci voted on patch set 2 V +1 12/09/25, 18:05 Core CI is happy: https://git.typo3.org/typo3/CI/cms/-/pipelines/91924

  6. Nikita Hovratov commented on patch set 2 12/09/25, 19:38

    Reply

    I'm surprised as well. No, you just revelead it with your added strict type in Extbase Parameter that something weird was going on. The parameters were wrong in v12 too, but didn't cause any trouble / PHP exceptions.
  7. Christian Kuhn commented on patch set 2 12/09/25, 21:30

    Reply

    i see. interesting. thx. i'll try to give the surrounding code a quick look soon in order to come up with a vote.
  8. Christian Kuhn commented on patch set 2 CR +1 12/09/25, 22:57

    +1 on reading for now. we should have another person having a detailed look here.
  9. Stefan Bürk commented on patch set 2 CR -1 12/10/25, 06:48

    Reviewing ...
  10. Stefan Bürk commented on patch set 2 12/10/25, 10:31

    Reply

    Keeping the -1 for now. Background: I had a short call with Nikita and a first try to solve this in another way (instead of moving the issue) did not worked out. I need to create a local test case and debug that further. Nikita expressed to me that he is not sure about the fix in general, albeit working for him as composer patch and that we take the time to investigate this deeper first instead if doing a more-or-less qick fix and possible shift the bug from one users to others.
  11. Nikita Hovratov uploaded patch set 3 CR +1 V +1 12/12/25, 22:36

  12. core-ci voted on patch set 3 V -1 12/12/25, 22:45 Core CI is not happy: https://git.typo3.org/typo3/CI/cms/-/pipelines/92166

  13. Nikita Hovratov uploaded patch set 4 V -1 12/12/25, 22:59

  14. Nikita Hovratov commented on patch set 4 12/12/25, 23:05

    Reply

    Hey @stefan@buerk.tech, I added a functional test for you, which shows the error, when you revert the third underscore. Having a second look at the debugger, it's really hard to fix it somewhere else, because the whole inflate/deflate logic is based on the arguments being separated by the `__` delimiter like `internal__argument`. So this always ends up being `internal____argument`, which will be interpreted as an indexed array `internal[0][argument]`. Another idea could be to encode double underscore values, when deflating and decode again when inflating. But of course this also has to be unique somewhow.
  15. core-ci voted on patch set 4 V +1 12/12/25, 23:06 Core CI is happy: https://git.typo3.org/typo3/CI/cms/-/pipelines/92167

  16. Garvin Hicking commented on patch set 4 CR +1 V +1 03/15/26, 13:13

    I don't feel too sure for 13.4 in this, but I think for main we should give this a go now before LTS. Thoughts?
  17. Nikita Hovratov commented on patch set 4 03/16/26, 08:27

    Reply

    Fine by me. I hope @stefan@buerk.tech can give this a pass as well.
  18. Garvin Hicking rebased onto patch set 5 CR +1 V +1 04/08/26, 17:23

  19. Garvin Hicking updated the commit message in patch set 6 04/08/26, 17:24

  20. Garvin Hicking commented on patch set 6 CR +1 V +1 04/08/26, 17:26

    Reply

    I scoped it to main. I believe this is something we should do in a .0 major release version, so now is the time - I could not come up with an idea where something actually breaks, but maybe cached (routing) URIs or so could misbehave, that's my main worry.
  21. Christian Kuhn commented on patch set 6 CR +1 V +1 04/08/26, 17:34

    Reply

    Done

    Reply

    agreed.
  22. core-ci voted on patch set 5 V +1 04/08/26, 17:34 Core CI is happy: https://git.typo3.org/typo3/CI/cms/-/pipelines/99785

  23. core-ci voted on patch set 6 V +1 04/08/26, 17:35 Core CI is happy: https://git.typo3.org/typo3/CI/cms/-/pipelines/99786

  24. Christian Kuhn voted on patch set 6 CR +2 V +2 04/08/26, 18:40

  25. Christian Kuhn merged patch set 7 04/08/26, 18:40