fix

Author: gatorsmile

sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala

@@ -552,6 +552,12 @@ object SQLConf {
     .booleanConf
     .createWithDefault(true)
 
+  val SUBQUERY_REUSE_ENABLED = buildConf("spark.sql.subquery.reuse")
+    .internal()
+    .doc("When true, the planner will try to find out duplicated subqueries and re-use them.")
+    .booleanConf
+    .createWithDefault(true)
+
   val STATE_STORE_PROVIDER_CLASS =
     buildConf("spark.sql.streaming.stateStore.providerClass")
       .internal()
@@ -932,6 +938,8 @@ class SQLConf extends Serializable with Logging {
 
   def exchangeReuseEnabled: Boolean = getConf(EXCHANGE_REUSE_ENABLED)
 
+  def subqueryReuseEnabled: Boolean = getConf(SUBQUERY_REUSE_ENABLED)
+
   def caseSensitiveAnalysis: Boolean = getConf(SQLConf.CASE_SENSITIVE)
 
   def constraintPropagationEnabled: Boolean = getConf(CONSTRAINT_PROPAGATION_ENABLED)

sql/core/src/main/scala/org/apache/spark/sql/execution/aggregate/HashAggregateExec.scala

@@ -25,6 +25,7 @@ import org.apache.spark.sql.catalyst.errors._
 import org.apache.spark.sql.catalyst.expressions._
 import org.apache.spark.sql.catalyst.expressions.aggregate._
 import org.apache.spark.sql.catalyst.expressions.codegen._
+import org.apache.spark.sql.catalyst.plans.QueryPlan
 import org.apache.spark.sql.catalyst.plans.physical._
 import org.apache.spark.sql.execution._
 import org.apache.spark.sql.execution.metric.{SQLMetric, SQLMetrics}

sql/core/src/main/scala/org/apache/spark/sql/execution/basicPhysicalOperators.scala

@@ -595,6 +595,9 @@ case class OutputFakerExec(output: Seq[Attribute], child: SparkPlan) extends Spa
  */
 case class SubqueryExec(name: String, child: SparkPlan) extends UnaryExecNode {
 
+  // Ignore this wrapper for canonicalizing.
+  override lazy val canonicalized: SparkPlan = child.canonicalized
+
   override lazy val metrics = Map(
     "dataSize" -> SQLMetrics.createMetric(sparkContext, "data size (bytes)"),
     "collectTime" -> SQLMetrics.createMetric(sparkContext, "time to collect (ms)"))

sql/core/src/main/scala/org/apache/spark/sql/execution/subquery.scala

@@ -156,7 +156,7 @@ case class PlanSubqueries(sparkSession: SparkSession) extends Rule[SparkPlan] {
 case class ReuseSubquery(conf: SQLConf) extends Rule[SparkPlan] {
 
   def apply(plan: SparkPlan): SparkPlan = {
-    if (!conf.exchangeReuseEnabled) {
+    if (!conf.subqueryReuseEnabled) {
       return plan
     }
     // Build a hash map using schema of subqueries to avoid O(N*N) sameResult calls.

sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala

@@ -23,9 +23,12 @@ import java.net.{MalformedURLException, URL}
 import java.sql.Timestamp
 import java.util.concurrent.atomic.AtomicBoolean
 
+import scala.collection.mutable.ArrayBuffer
+
 import org.apache.spark.{AccumulatorSuite, SparkException}
 import org.apache.spark.scheduler.{SparkListener, SparkListenerJobStart}
 import org.apache.spark.sql.catalyst.util.StringUtils
+import org.apache.spark.sql.execution.{ScalarSubquery, SubqueryExec}
 import org.apache.spark.sql.execution.aggregate
 import org.apache.spark.sql.execution.joins.{BroadcastHashJoinExec, CartesianProductExec, SortMergeJoinExec}
 import org.apache.spark.sql.functions._
@@ -700,6 +703,33 @@ class SQLQuerySuite extends QueryTest with SharedSQLContext {
         row => Seq.fill(16)(Row.merge(row, row))).collect().toSeq)
   }
 
+  test("Verify spark.sql.subquery.reuse") {
+    Seq("true", "false").foreach { reuse =>
+      withSQLConf(SQLConf.SUBQUERY_REUSE_ENABLED.key -> reuse) {
+        val df = sql(
+          """
+            |SELECT key, (SELECT avg(key) FROM testData)
+            |FROM testData
+            |WHERE key > (SELECT avg(key) FROM testData)
+          """.stripMargin)
+        val subqueries = ArrayBuffer.empty[SubqueryExec]
+        df.queryExecution.executedPlan.transformAllExpressions {
+          case s @ ScalarSubquery(plan: SubqueryExec, _) =>
+            subqueries += plan
+            s
+        }
+
+        assert(subqueries.size == 2, "Two ScalarSubquery are expected in the plan")
+
+        if (reuse.toBoolean) {
+          assert(subqueries.distinct.size == 1, "Only one ScalarSubquery exists in the plan")
+        } else {
+          assert(subqueries.distinct.size == 2, "Reuse is not expected")
+        }
+      }
+    }
+  }
+
   test("cartesian product join") {
     withSQLConf(SQLConf.CROSS_JOINS_ENABLED.key -> "true") {
       checkAnswer(