aio-libs / multidict / 36545853476

87%

push

github

Honor --impl in the driver's standalone --self-check (#1637)

<!-- Thank you for your contribution! -->

## What do these changes do?

`benchmarks/callgrind_driver.py --self-check` on its own checked every
implementation, ignoring `--impl`. On a pure-Python install it failed on
`multidict_c` with the hint "pick an importable implementation with
--impl", which could not help in that mode (Greptile on #1624). It now
checks every operation of the implementation `--impl` names, or of all
of them without it, as before.

## Are there changes in behavior for the user?

No, benchmark tooling only.

## Is it a substantial burden for the maintainers to support this?

No.

## Related issue number

Follows #1624. Supersedes #1632, which changed the hint instead and left
it wrong for a failing pure-Python implementation.

## Checklist

- [x] I think the code is well written
- [x] Unit tests for the changes exist
- [x] Documentation reflects the changes
- [ ] If you provide code modification, please add yourself to
`CONTRIBUTORS.txt` (N/A)
- [x] Add a new news fragment into the `CHANGES/` folder
- [x] `make doc-spelling` passes and any new technical words are added
to `docs/spelling_wordlist.txt`

<details>
<summary>Agent run details (optional, for reviewers)</summary>

`test_main_self_check_checks_every_cell` is parametrized over no
`--impl` and `--impl multidict_py`; `tests/test_callgrind_driver.py`
passes (6) with every line covered. On a `MULTIDICT_NO_EXTENSIONS=1`
install, `--self-check --impl multidict_py` passes (30 cells);
`--self-check` alone still fails on `multidict_c`, now with a hint that
works.

Lint: pre-commit clean; `make doc-spelling` passes.
</details>

Drafted with Claude Code (Claude Opus 5.5); reviewed by @asvetlov.

788 of 1576 branches covered (50.0%)

Branch coverage included in aggregate %.

4 of 5 new or added lines in 1 file covered. (80.0%)

8961 of 9638 relevant lines covered (92.98%)

1.86 hits per line