cloud-fan commented on a change in pull request #35360:
URL: https://github.com/apache/spark/pull/35360#discussion_r801672096
##########
File path:
sql/catalyst/src/main/java/org/apache/spark/sql/connector/read/SupportsPushDownAggregates.java
##########
@@ -47,6 +47,13 @@
@Evolving
public interface SupportsPushDownAggregates extends ScanBuilder {
+ /**
+ * Whether enable aggregation push-down.
+ *
+ * @return true if enable aggregation push-down, false otherwise.
+ */
+ default boolean enablePushAggregation() { return true; }
Review comment:
> The current implement do a lot work(e.g. collect aggregates, normalize
expressions, translate aggregation, split AVG into SUM and COUNT and so on).
Have you measured the actual performance penalty? I'd strongly be against
adding a new API just for a tiny performance gain.
--
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]
