jayzhan211 commented on PR #8137: URL: https://github.com/apache/arrow-datafusion/pull/8137#issuecomment-1807282038
> Thanks @jayzhan211 Numbers for two rows is definitely promising. However I'm not sure I'm following what is a single row or two row. perhaps you can attach a benchmark code to make it more clear? Single row is like [[1,2,3]] or ListArray( Int64Array(1,2,3) ). Two row is like [[1,2,3], [4,5,6]] or ListArray (Int64Array(1,2,3), (4,5,6)) which is the target we are trying to optimized. We dont have to `compute::concat` each rows at the end instead we `extend` each row we need while iterating. This is why we can see `Tow row` in this change outperform the original 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
