costin opened a new pull request, #16355: URL: https://github.com/apache/lucene/pull/16355
CharacterUtils.toLowerCase() runs the full Unicode codepoint path for every character, even when the token is already all-lowercase ASCII, common case after English tokenization. The PR introduces a fast path that uses direct arithmetic for ASCII-only tokens instead of the generic Unicode approach. ### Benchmark AMD EPYC 7R32 (c5a.2xlarge), JDK 25, 1000 tokens (90% lowercase ASCII, 8% mixed-case, 2% non-ASCII). 3 forks, 10 iters × 1s. | method | throughput (ops/μs) | vs baseline | |---|---:|---:| | baseline (current) | 0.056 ± 0.001 | — | | asciiFastPath | 0.122 ± 0.001 | 2.18x | -- 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]
