JanKaul commented on code in PR #15862:
URL: https://github.com/apache/datafusion/pull/15862#discussion_r2072332676
##########
datafusion/expr/src/logical_plan/dml.rs:
##########
@@ -241,3 +241,10 @@ fn make_count_schema() -> DFSchemaRef {
.unwrap(),
)
}
+
Review Comment:
Actually the join makes sense. You'll probably always need to perform it.
I think one issue we're going to run into is that you will need to somehow
fork the join node. You will need a stream of record batches for the matching
flags and then you will need to reuse the join node to get a stream of record
batches for the not matching flags.
This is currently not well supported with datafusions execution model. When
you currently use the results of a node you can't reuse them anywhere else.
The case expression is a bit unclear to me. But maybe I just have to try to
understand it better. I thought it would be easier to keep the match clauses
separate. So that the later implementation can handle them more easily.
--
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]
