gh-152228: Fix an assertion failure in str.replace under a limited memory case

[sobolevn]

Because test was simple enough and was always 0 as the memory limit, I decided to include it.

• Issue: str.replace can crash under memory pressure #152228

[sobolevn]

Created #152236 to track Py_TRACE_REFS skips. There are several tests that don't have that.

[vstinner]

I wouldn't call it a "crash". It's "only" an assertion failure when Python is built in debug mode.

[vstinner]

LGTM.

[vstinner]

Strange, CI tests pass, but test_str fails when I download the PR locally:

$ ./python -m test test_str -v
...
test test_str crashed -- Traceback (most recent call last):
  ...
  File "/home/vstinner/python/main/Lib/test/test_str.py", line 610, in StrTest
    @support.nomemtest
     ^^^^^^^^^^^^^^^^^
AttributeError: module 'test.support' has no attribute 'nomemtest'. Did you mean '.bigmemtest' instead of '.nomemtest'?

Do it work on your machine? The "Squash and merge" should work around the issue anyway.

[sobolevn]

Looks like a forgot to push the merge commit. Did it now :)
Thanks a lot for the reviews!

[vstinner]

Looks like a forgot to push the merge commit. Did it now :)

But it's strange that tests passed on the CI (without the merge commit).

[sobolevn]

I copied this exact branch, added assert False to the test, executed it:

test_replace_oom failed. It worked correctly for me :)

[miss-islington-app]

Thanks @sobolevn for the PR 🌮🎉.. I'm working now to backport this PR to: 3.13, 3.14, 3.15.
🐍🍒⛏🤖 I'm not a witch! I'm not a witch!

[bedevere-app]

GH-152615 is a backport of this pull request to the 3.15 branch.

[bedevere-app]

GH-152616 is a backport of this pull request to the 3.14 branch.

[bedevere-app]

GH-152617 is a backport of this pull request to the 3.13 branch.