[GitHub] [spark] ulysses-you commented on a diff in pull request #36953: [SPARK-39551][SQL] Add AQE invalid plan check

[GitBox <gi...@apache.org>]

ulysses-you commented on code in PR #36953:
URL: https://github.com/apache/spark/pull/36953#discussion_r903303288


##########
sql/core/src/main/scala/org/apache/spark/sql/execution/adaptive/ValidateSparkPlan.scala:
##########
@@ -0,0 +1,68 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.sql.execution.adaptive
+
+import org.apache.spark.sql.catalyst.optimizer.{BuildLeft, BuildRight}
+import org.apache.spark.sql.catalyst.rules.Rule
+import org.apache.spark.sql.execution.SparkPlan
+import org.apache.spark.sql.execution.joins.{BroadcastHashJoinExec, BroadcastNestedLoopJoinExec}
+
+/**
+ * Detects invalid physical plans generated by AQE replanning and throws `InvalidAQEPlanException`
+ * if such plans are detected. This rule should be called after EnsureRequirements where all
+ * necessary Exchange nodes are added.
+ */
+object ValidateSparkPlan extends Rule[SparkPlan] {
+
+  def apply(plan: SparkPlan): SparkPlan = {
+    validate(plan)
+    plan
+  }
+
+  /**
+   * Validate that the plan satisfies the following condition:
+   * - BroadcastQueryStage only appears as the immediate child and the build side of a broadcast
+   *   hash join or broadcast nested loop join.
+   */
+  private def validate(plan: SparkPlan): Unit = plan match {
+    case b: BroadcastHashJoinExec =>
+      val (buildPlan, probePlan) = b.buildSide match {
+        case BuildLeft => (b.left, b.right)
+        case BuildRight => (b.right, b.left)
+      }
+      if (!buildPlan.isInstanceOf[BroadcastQueryStageExec]) {

Review Comment:
   if the BroadcastHashJoinExec is from a sort merge join, it's direct child should be BroadcastExchangeExec ? for example, two joins, the first join is BroadcastHashJoinExec and the second join is sort merge join at first then after re-optimze it changes to broadcast join.
   
   Or we can move this rule after the `createQueryStages`



-- 
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: reviews-unsubscribe@spark.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org