Skip to content

perf: remove redundant .lower() calls in Headers#3735

Open
shindonghwi wants to merge 1 commit intoencode:masterfrom
shindonghwi:perf/remove-redundant-lower-calls
Open

perf: remove redundant .lower() calls in Headers#3735
shindonghwi wants to merge 1 commit intoencode:masterfrom
shindonghwi:perf/remove-redundant-lower-calls

Conversation

@shindonghwi
Copy link

@shindonghwi shindonghwi commented Dec 19, 2025 •
edited
Loading

`Headers._list` stores tuples as `(raw_key, lowercase_key, value)`.

In `get_list()` and `delitem()`, calling `.lower()` on `item_key` is redundant since it's already lowercase.

# Before
if item_key.lower() == get_header_key

# After  
if item_key == get_header_key

Benchmark

# 10 headers, 100k iterations
_list = [(key, key.lower(), b'value') for key in headers]
get_header_key = b'x-header-5'

# Original
[v for _, k, v in _list if k.lower() == get_header_key]  # 0.0647s

# Fixed
[v for _, k, v in _list if k == get_header_key]          # 0.0437s

Result: 32.5% faster (1.48x speedup)

@shindonghwi
perf: remove redundant .lower() calls in Headers
4777295 
The _list stores (raw_key, lowercase_key, value) tuples, so the
second element is already lowercase. Calling .lower() on it again
is unnecessary and wastes CPU cycles.

Benchmark shows ~1.6x speedup for header lookups.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@shindonghwi