cloud-fan commented on code in PR #42317:
URL: https://github.com/apache/spark/pull/42317#discussion_r1356530715
##########
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/InjectRuntimeFilter.scala:
##########
@@ -118,17 +122,34 @@ object InjectRuntimeFilter extends Rule[LogicalPlan] with
PredicateHelper with J
* filter should be selective and the filter condition (including
expressions in the child
* plan referenced by the filter condition) should be a simple expression,
so that we do
* not add a subquery that might have an expensive computation.
Review Comment:
let's add more doc to explain why we can extract a sub-plan to filter the
application side.
```
The extracted sub-plan should produce a superset of the entire creation side
output data, so that
it's still correct to use the sub-plan to build the runtime filter to prune
the application side.
```
--
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]
