maartenbreddels commented on pull request #7656: URL: https://github.com/apache/arrow/pull/7656#issuecomment-655452317
The benchmark ran `isalpha`, which was a bad idea (high probability for a false on the first character), `isalnum` is a better test case, with characters in the range `A`-'z' (including some non-alpnum characters). Before: ``` IsAlphaNumericUnicode_median 68083381 ns 68066554 ns 3 bytes_per_second=232.786M/s items_per_second=15.4052M/s ``` After: ``` IsAlphaNumericUnicode_median 44940488 ns 44931459 ns 3 bytes_per_second=352.647M/s items_per_second=23.3372M/s ``` This is still 3x slower than what I have in Vaex, so I guess the compiler doesn't like my code. ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
