perf: optimize regexp_count to avoid String allocation when start position is provided #19553
+133
−2
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ition is provided Replace `.chars().skip().collect::<String>()` with zero-copy string slicing using `char_indices()` to find the byte offset, then slice with `&value[byte_offset..]`. This eliminates unnecessary String allocation per row when a start position is specified. Changes: - Use char_indices().nth() to find byte offset for start position (1-based) - Use string slicing &value[byte_offset..] instead of collecting chars - Added benchmark to measure performance improvements Optimization: - Before: Allocated new String via .collect() for each row with start position - After: Uses zero-copy string slice Benchmark results: - size=1024, str_len=32: 96.361 µs -> 41.458 µs (57.0% faster, 2.3x speedup) - size=1024, str_len=128: 210.16 µs -> 56.064 µs (73.3% faster, 3.7x speedup) - size=4096, str_len=32: 376.90 µs -> 162.98 µs (56.8% faster, 2.3x speedup) - size=4096, str_len=128: 855.68 µs -> 263.61 µs (69.2% faster, 3.2x speedup) The optimization shows greater improvements for longer strings (up to 73% faster) since string slicing is O(1) regardless of length, while the previous approach had allocation costs that grew with string length.
viirya
force-pushed
the
regexp_count_optimize
branch
from
8f465d6 to
dbdc19c
Compare
comphead
reviewed
Dec 30, 2025
| .unwrap_or(value.len()); | ||
|
|
||
| // Use string slicing instead of collecting chars into a new String | ||
| let find_slice = &value[byte_offset..]; |
Contributor
There was a problem hiding this comment.
Thanks @viirya a quick question as this PR refers to the string by offset directly, would be that UTF compliant?
Member
Author
There was a problem hiding this comment.
Yes. char_indices is the same as chars working on Unicode Scalar Value.
comphead
approved these changes
Dec 30, 2025
Member
Author
|
Thanks @comphead |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Which issue does this PR close?
Rationale for this change
What changes are included in this PR?
Replace
.chars().skip().collect::<String>()with zero-copy string slicing usingchar_indices()to find the byte offset, then slice with&value[byte_offset..].This eliminates unnecessary String allocation per row when a start position is specified.
Changes:
Optimization:
Benchmark results:
The optimization shows greater improvements for longer strings (up to 73% faster) since string slicing is O(1) regardless of length, while the previous approach had allocation costs that grew with string length.
Are these changes tested?
Are there any user-facing changes?