Github user liancheng commented on a diff in the pull request: https://github.com/apache/spark/pull/8652#discussion_r42286246 --- Diff: sql/core/src/main/scala/org/apache/spark/sql/execution/SparkStrategies.scala --- @@ -268,6 +268,27 @@ private[sql] abstract class SparkStrategies extends QueryPlanner[SparkPlan] { object CartesianProduct extends Strategy { def apply(plan: LogicalPlan): Seq[SparkPlan] = plan match { + // Not like the equal-join, BroadcastNestedLoopJoin doesn't support condition + // for cartesian join, as in cartesian join, probably, the records satisfy the + // condition, but exists in another partition of the large table, so we may not able + // to eliminate the duplicates. + case logical.Join( + CanBroadcast(left), right, joinType @ (FullOuter | LeftOuter | RightOuter), None) => + execution.joins.BroadcastNestedLoopJoin( + planLater(left), planLater(right), joins.BuildLeft, joinType, None) :: Nil + case logical.Join( + left, CanBroadcast(right), joinType @ (FullOuter | LeftOuter | RightOuter), None) => + execution.joins.BroadcastNestedLoopJoin( + planLater(left), planLater(right), joins.BuildRight, joinType, None) :: Nil + // Since BroadCastNestedLoopJoin supports condition already, we simply passed it down. --- End diff -- Indentation is off
--- 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 infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org