mightsleep opened a new pull request, #26413:
URL: https://github.com/apache/datafusion/pull/26413
## Which issue does this PR close?
N/A. Prepares a follow-up that speeds up the non-ASCII path of left and
right; the cases go in first so the benchmark bot has a baseline.
## Rationale for this change
The left_right benchmark only uses ASCII inputs. Since #26039 those take the
ASCII fast path, so the path that scans for character boundaries in non-ASCII
strings is not measured at all.
## What changes are included in this PR?
- A generator for Cyrillic strings (two bytes per letter, spaces between
words), in characters.
- Three non-ASCII cases for both Utf8 and Utf8View: short_result (n = 5, 1
to 32 characters), long_result (n, negative_n (n = -100, 64 to256 characters).
## What is the testing strategy for this PR?
Benchmark only; cargo bench -p datafusion-functions --bench
unicode_expressions --no-run builds it.
## Are there any user-facing changes?
No.
--
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]
