aio-libs / multidict / 35794162320

87%

push

github

Add CIMultiDict benchmarks for plain str keys (#1521)

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

## What do these changes do?

Adds six CodSpeed benchmarks for `CIMultiDict` keyed by a plain `str`.

Every `CIMultiDict` benchmark in the suite today uses either `istr`
keys,
which serve the identity straight from the cached canonical form, or
digit-only `str` keys such as `str(i)`, which are already lowercase. So
the
case-insensitive lowering that a mixed-case key goes through on every
single
operation is currently unmeasured.

The new benchmarks use a list of 200 realistic HTTP header names, in
mixed-case and lowercase spellings, plus a non-ASCII control:
`insert_str_mixed_case`, `insert_str_lower_case`, `add_str_mixed_case`,
`fetch_str_mixed_case`, `fetch_str_lower_case` and
`fetch_str_non_ascii`.

This is groundwork for an optimisation of `_ci_key_to_identity()`;
landing the
benchmarks separately gives that change a real baseline on `master` to
compare
against, rather than a report on names that did not exist on the base
commit.

## Are there changes in behavior for the user?

No, this only adds benchmarks.

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

No, they follow the shape of the surrounding benchmarks and run on both
backends through the existing `case_insensitive_multidict_class`
fixture.

## Related issue number

N/A

## Checklist

- [x] I think the code is well written
- [x] Unit tests for the changes exist - N/A, this is a benchmark-only
change
- [x] Documentation reflects the changes - N/A
- [ ] If you provide code modification, please add yourself to
`CONTRIBUTORS.txt` - N/A, no such file in this repo
- [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`

Drafted with Claude Code (Claude Opus 5); reviewed by @asvetlov.

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

Tests: `pytest -q t... (continued)

765 of 1530 branches covered (50.0%)

Branch coverage included in aggregate %.

69 of 69 new or added lines in 1 file covered. (100.0%)

6860 of 7147 relevant lines covered (95.98%)

1.92 hits per line