Github user gatorsmile commented on a diff in the pull request:
https://github.com/apache/spark/pull/18692#discussion_r153060560
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/joins.scala
---
@@ -152,3 +152,99 @@ object EliminateOuterJoin extends Rule[LogicalPlan]
with PredicateHelper {
if (j.joinType == newJoinType) f else Filter(condition,
j.copy(joinType = newJoinType))
}
}
+
+/**
+ * A rule that eliminates CROSS joins by inferring join conditions from
propagated constraints.
+ *
+ * The optimization is applicable only to CROSS joins. For other join
types, adding inferred join
+ * conditions would potentially shuffle children as child node's
partitioning won't satisfy the JOIN
+ * node's requirements which otherwise could have.
+ *
+ * For instance, if there is a CROSS join, where the left relation has 'a
= 1' and the right
+ * relation has 'b = 1', the rule infers 'a = b' as a join predicate.
+ */
+object EliminateCrossJoin extends Rule[LogicalPlan] with PredicateHelper {
+
+ def apply(plan: LogicalPlan): LogicalPlan = {
+ if (SQLConf.get.constraintPropagationEnabled) {
+ eliminateCrossJoin(plan)
+ } else {
+ plan
+ }
+ }
+
+ private def eliminateCrossJoin(plan: LogicalPlan): LogicalPlan = plan
transform {
+ case join@Join(leftPlan, rightPlan, Cross, None) =>
--- End diff --
Nit: `join@Join` -> `join @ Join`
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
