Ted-Jiang commented on PR #4616: URL: https://github.com/apache/arrow-datafusion/pull/4616#issuecomment-1352516495
> I think in the long run breaking the update_batch and retract_batch API makes code hard to maintain. Yes, I agree with this. Now window function with agg reuse the aggExpr, if we try to imply `retract_batch` for min max, i think it will also break a lot of api for avoiding effect the efficiency of non custom window frames situation. So I prefer keep this alone, is there any solution🤔 > I think this data structure is better for our use case Yes, you are right, from the paper why need `stores all batch at the beginning ` i think it will make `inter-partition parallelism` (but not in this pr 😂) -- 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: github-unsubscr...@arrow.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
