Dandandan commented on PR #5912: URL: https://github.com/apache/arrow-datafusion/pull/5912#issuecomment-1501509563
> > One way to accomplish it would be differentating window and conventional aggregations, or maybe it could be handled inside the window function code itself 🤔 > > I suspect this will result in more complexity than what we started with. But one can never be sure without trying and playing around. Playing a bit with the code, my feeling is that usage of the accumulator in window functions is suboptimal, because of producing a `ScalarValue` for each output. I think when moving to more specialized kernels for (moving) sum/avg/count we could remove the usage of `SumAccumulator` and remove the `count` from it. -- 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]
