neilconway opened a new pull request, #21064: URL: https://github.com/apache/datafusion/pull/21064
## Which issue does this PR close? - Closes #21039 . ## Rationale for this change For short strings that are stored inline in a `Utf8View`, we can hash the string's value directly, without materializing a `&str`, and then add the hash value to HyperLogLog directly. This improves performance by ~40%. ## What changes are included in this PR? * Add benchmark for `approx_distinct` on short strings * Add `add_hashed` API to `HyperLogLog` * Rename `SEED` to `HLL_HASH_STATE` and make it `pub(crate)` * Optimize `approx_distinct` on short strings as described above. ## Are these changes tested? Yes. ## 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]
