Github user fhueske commented on a diff in the pull request:
https://github.com/apache/flink/pull/1567#discussion_r51731066
--- Diff:
flink-libraries/flink-table/src/main/scala/org/apache/flink/api/table/plan/rules/dataset/DataSetJoinRule.scala
---
@@ -39,6 +46,10 @@ class DataSetJoinRule
val convLeft: RelNode = RelOptRule.convert(join.getInput(0),
DataSetConvention.INSTANCE)
val convRight: RelNode = RelOptRule.convert(join.getInput(1),
DataSetConvention.INSTANCE)
+ val joinKeys = getJoinKeys(join)
+
+ // There would be a FlinkProject after FlinkJoin to handle the output
fields afterward join,
+ // so we do not need JoinFunction here by now.
--- End diff --
We need the `JoinFunction` to evaluate all non-equality join predicates.
For instance if we have a join predicate such as `t1.a == t2.a AND t1.b <
t2.b`, we would evaluate the `t1.a == t2.a` part by the joinKeys of the
JoinOperator and the `t1.b < t2.b` as part of the `JoinFunction`.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
