hsiang-c commented on code in PR #4727: URL: https://github.com/apache/datafusion-comet/pull/4727#discussion_r3476495208
########## spark/src/main/scala/org/apache/spark/sql/comet/operators.scala: ########## @@ -1892,19 +1892,19 @@ object CometObjectHashAggregateExec } /** - * For Partial mode aggregates containing TypedImperativeAggregate functions (like CollectSet), - * the Spark-side output declares buffer columns as BinaryType (since Spark serializes state to - * binary). However, the native Comet aggregate produces the actual state type (e.g., - * ArrayType(elementType) for CollectSet). This method corrects the output schema to match the - * native state types so the shuffle exchange schema is consistent with the actual data. + * For intermediate aggregates containing TypedImperativeAggregate functions (like CollectSet or Review Comment: 👍 Could we add a few SQL file tests: https://datafusion.apache.org/comet/contributor-guide/sql-file-tests.html#sql-file-tests, thank you! -- 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]
