andygrove opened a new pull request, #23540:
URL: https://github.com/apache/datafusion/pull/23540
> This PR was created by an LLM as a draft PR. I will mark it as ready for
review after human review.
## Which issue does this PR close?
N/A — autonomous exploratory PR.
## Rationale for this change
Removed per-batch default-array/Vec allocations, memoized the compiled regex
across rows to skip HashMap hashing for literal patterns, and skipped the O(n)
chars() scan on the default start=1 path.
## What changes are included in this PR?
Removed per-batch default-array/Vec allocations, memoized the compiled regex
across rows to skip HashMap hashing for literal patterns, and skipped the O(n)
chars() scan on the default start=1 path.
## Are these changes tested?
Correctness: unit tests + seeded differential fuzz (bit-identical Arrow
output vs `main`).
Benchmark (criterion):
- regexp_instr_with_start [size=1024, str_len=32]: 40.073% faster (base
41127ns -> cand 24646ns)
- regexp_instr_with_start [size=1024, str_len=128]: 44.498% faster (base
48215ns -> cand 26760ns)
- regexp_instr_no_start [size=1024, str_len=128]: 47.253% faster (base
49244ns -> cand 25974ns)
- regexp_instr_no_start [size=1024, str_len=32]: 42.18% faster (base 40495ns
-> cand 23414ns)
Full criterion output:
```text
regexp_instr_no_start [size=1024, str_len=32]
time: [23.389 µs 23.479 µs 23.645 µs]
change: [−42.300% −42.180% −42.004%] (p = 0.00 <
0.05)
Performance has improved.
Found 9 outliers among 100 measurements (9.00%)
7 (7.00%) low mild
2 (2.00%) high severe
regexp_instr_with_start [size=1024, str_len=32]
time: [24.617 µs 24.635 µs 24.659 µs]
change: [−40.118% −40.073% −40.026%] (p = 0.00 <
0.05)
Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high severe
regexp_instr_no_start [size=1024, str_len=128]
time: [25.756 µs 25.791 µs 25.830 µs]
change: [−47.394% −47.253% −47.076%] (p = 0.00 <
0.05)
Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high severe
regexp_instr_with_start [size=1024, str_len=128]
time: [26.645 µs 26.666 µs 26.693 µs]
change: [−44.577% −44.498% −44.417%] (p = 0.00 <
0.05)
Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high mild
```
## Are there any user-facing changes?
<!--
If there are user-facing changes then we may require documentation to be
updated before approving the PR.
-->
<!--
If there are any breaking changes to public APIs, please add the `api
change` label.
-->
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
