gh-124722: Fix leak in test_detach_materialized_dict_no_memory

[sobolevn]

The leak was in setting _testcapi.set_nomemory(0, 1), in other places it is only used together with assert_python_ok

I also marked this test as cpython_only and I also skip it if _testcapi is not present.

Local run:

» ./python.exe -m test test_class -m test_detach_materialized_dict_no_memory -R 3:3 Using random seed: 365282797 0:00:00 load avg: 1.55 Run 1 test sequentially in a single process 0:00:00 load avg: 1.55 [1/1] test_class beginning 6 repetitions. Showing number of leaks (. for 0 or less, X for 10 or more) 123:456 XX. ... == Tests result: SUCCESS == 1 test OK. Total duration: 403 ms Total tests: run=1 (filtered) Total test files: run=1/1 (filtered) Result: SUCCESS 

• Issue: test_class.TestInlineValues.test_detach_materialized_dict_no_memory leaks references #124722

[ZeroIntensity]

I think this looks fine, but I'm worried about what problems that _testcapi.set_nomemory is causing. Could you elaborate on that? (This seems like more of a bandaid than an actual fix for that function.)

[Eclips4]

LGTM! I thought of the same approach.

[sobolevn]

@ZeroIntensity please, take a look at fm_setup_hooks, probably something there is not cleaned up.

The more complex fix would be:

• make sure that fm_remove_hooks completely disables all modifications done in fm_setup_hooks
• create a context manager to automatically clean up this context in tests

But, this change would be much more complex, that's why I went with the simplier (and already existing) fix instead.

[ZeroIntensity]

Yeah, it's good to get this in sooner. I'll look into fixing set_nomemory, but it's odd because as far as I can see fm_setup_hooks makes no additional allocations, so I'm not exactly sure what could be getting leaked.

[zware]

!buildbot .RHEL8 Refleak.

[bedevere-bot]

🤖 New build scheduled with the buildbot fleet by @zware for commit e7a409b 🤖

The command will test the builders whose names match following regular expression: .*RHEL8 Refleak.*

The builders matched are:

• s390x RHEL8 Refleaks PR
• AMD64 RHEL8 Refleaks PR
• aarch64 RHEL8 Refleaks PR
• PPC64LE RHEL8 Refleaks PR

[zware]

The buildbots look happy with this, so I'm going ahead with the merge to get them cleaned up. @markshannon, it could still use a post-merge OK from you to confirm that it's still testing what you wanted to test.

[miss-islington-app]

Thanks @sobolevn for the PR, and @zware for merging it 🌮🎉.. I'm working now to backport this PR to: 3.13.
🐍🍒⛏🤖

[bedevere-app]

GH-124777 is a backport of this pull request to the 3.13 branch.