[GitHub] [arrow-datafusion] poonai commented on a diff in pull request #3473: add `PhysicalExpr` to support, Selective Aggregates feature using `FILTER` clause

[GitBox <gi...@apache.org>]

poonai commented on code in PR #3473:
URL: https://github.com/apache/arrow-datafusion/pull/3473#discussion_r1012905307


##########
datafusion/core/src/physical_plan/aggregates/hash.rs:
##########
@@ -200,11 +200,6 @@ fn group_aggregate_batch(
     // evaluate the grouping expressions
     let group_by_values = evaluate_group_by(group_by, &batch)?;
 
-    // evaluate the aggregation expressions.
-    // We could evaluate them after the `take`, but since we need to evaluate all
-    // of them anyways, it is more performant to do it while they are together.
-    let aggr_input_values = evaluate_many(aggregate_expressions, &batch)?;

Review Comment:
   I’ve checked it via running benchmark. I do not see any perf issue. 
   
   Maybe I’m wrong :( benchmark is tricky  



-- 
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-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org