[SPARK-21619][SQL] Fail the execution of canonicalized plans explicitly

## What changes were proposed in this pull request?
Canonicalized plans are not supposed to be executed. I ran into a case in which there's some code that accidentally calls execute on a canonicalized plan. This patch throws a more explicit exception when that happens.

## How was this patch tested?
Added a test case in SparkPlanSuite.

Author: Reynold Xin <[email protected]>

Closes #18828 from rxin/SPARK-21619.

Author: rxin

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/catalog/interface.scala

@@ -438,7 +438,7 @@ case class HiveTableRelation(
 
   def isPartitioned: Boolean = partitionCols.nonEmpty
 
-  override lazy val canonicalized: HiveTableRelation = copy(
+  override def doCanonicalize(): HiveTableRelation = copy(
     tableMeta = tableMeta.copy(
       storage = CatalogStorageFormat.empty,
       createTime = -1
@@ -448,7 +448,8 @@ case class HiveTableRelation(
     },
     partitionCols = partitionCols.zipWithIndex.map {
       case (attr, index) => attr.withExprId(ExprId(index + dataCols.length))
-    })
+    }
+  )
 
   override def computeStats(): Statistics = {
     tableMeta.stats.map(_.toPlanStats(output)).getOrElse {

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/QueryPlan.scala

@@ -180,6 +180,15 @@ abstract class QueryPlan[PlanType <: QueryPlan[PlanType]] extends TreeNode[PlanT
 
   override protected def innerChildren: Seq[QueryPlan[_]] = subqueries
 
+  /**
+   * A private mutable variable to indicate whether this plan is the result of canonicalization.
+   * This is used solely for making sure we wouldn't execute a canonicalized plan.
+   * See [[canonicalized]] on how this is set.
+   */
+  @transient private var _isCanonicalizedPlan: Boolean = false
+
+  protected def isCanonicalizedPlan: Boolean = _isCanonicalizedPlan
+
   /**
    * Returns a plan where a best effort attempt has been made to transform `this` in a way
    * that preserves the result but removes cosmetic variations (case sensitivity, ordering for
@@ -188,10 +197,24 @@ abstract class QueryPlan[PlanType <: QueryPlan[PlanType]] extends TreeNode[PlanT
    * Plans where `this.canonicalized == other.canonicalized` will always evaluate to the same
    * result.
    *
-   * Some nodes should overwrite this to provide proper canonicalize logic, but they should remove
-   * expressions cosmetic variations themselves.
+   * Plan nodes that require special canonicalization should override [[doCanonicalize()]].
+   * They should remove expressions cosmetic variations themselves.
+   */
+  @transient final lazy val canonicalized: PlanType = {
+    var plan = doCanonicalize()
+    // If the plan has not been changed due to canonicalization, make a copy of it so we don't
+    // mutate the original plan's _isCanonicalizedPlan flag.
+    if (plan eq this) {
+      plan = plan.makeCopy(plan.mapProductIterator(x => x.asInstanceOf[AnyRef]))
+    }
+    plan._isCanonicalizedPlan = true
+    plan
+  }
+
+  /**
+   * Defines how the canonicalization should work for the current plan.
    */
-  lazy val canonicalized: PlanType = {
+  protected def doCanonicalize(): PlanType = {
     val canonicalizedChildren = children.map(_.canonicalized)
     var id = -1
     mapExpressions {
@@ -213,7 +236,6 @@ abstract class QueryPlan[PlanType <: QueryPlan[PlanType]] extends TreeNode[PlanT
     }.withNewChildren(canonicalizedChildren)
   }
 
-
   /**
    * Returns true when the given query plan will return the same results as this query plan.
    *

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/basicLogicalOperators.scala

@@ -760,7 +760,7 @@ case class SubqueryAlias(
     child: LogicalPlan)
   extends UnaryNode {
 
-  override lazy val canonicalized: LogicalPlan = child.canonicalized
+  override def doCanonicalize(): LogicalPlan = child.canonicalized
 
   override def output: Seq[Attribute] = child.output.map(_.withQualifier(Some(alias)))
 }

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/hints.scala

@@ -41,7 +41,7 @@ case class ResolvedHint(child: LogicalPlan, hints: HintInfo = HintInfo())
 
   override def output: Seq[Attribute] = child.output
 
-  override lazy val canonicalized: LogicalPlan = child.canonicalized
+  override def doCanonicalize(): LogicalPlan = child.canonicalized
 }
 
 

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

@@ -139,7 +139,7 @@ case class RowDataSourceScanExec(
   }
 
   // Don't care about `rdd` and `tableIdentifier` when canonicalizing.
-  override lazy val canonicalized: SparkPlan =
+  override def doCanonicalize(): SparkPlan =
     copy(
       fullOutput.map(QueryPlan.normalizeExprId(_, fullOutput)),
       rdd = null,
@@ -522,7 +522,7 @@ case class FileSourceScanExec(
     }
   }
 
-  override lazy val canonicalized: FileSourceScanExec = {
+  override def doCanonicalize(): FileSourceScanExec = {
     FileSourceScanExec(
       relation,
       output.map(QueryPlan.normalizeExprId(_, output)),

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

@@ -111,6 +111,9 @@ abstract class SparkPlan extends QueryPlan[SparkPlan] with Logging with Serializ
    * Concrete implementations of SparkPlan should override `doExecute`.
    */
   final def execute(): RDD[InternalRow] = executeQuery {
+    if (isCanonicalizedPlan) {
+      throw new IllegalStateException("A canonicalized plan is not supposed to be executed.")
+    }
     doExecute()
   }
 
@@ -121,6 +124,9 @@ abstract class SparkPlan extends QueryPlan[SparkPlan] with Logging with Serializ
    * Concrete implementations of SparkPlan should override `doExecuteBroadcast`.
    */
   final def executeBroadcast[T](): broadcast.Broadcast[T] = executeQuery {
+    if (isCanonicalizedPlan) {
+      throw new IllegalStateException("A canonicalized plan is not supposed to be executed.")
+    }
     doExecuteBroadcast()
   }
 

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

@@ -286,7 +286,7 @@ class SparkSqlAstBuilder(conf: SQLConf) extends AstBuilder(conf) {
    * Create a [[ClearCacheCommand]] logical plan.
    */
   override def visitClearCache(ctx: ClearCacheContext): LogicalPlan = withOrigin(ctx) {
-    ClearCacheCommand
+    ClearCacheCommand()
   }
 
   /**

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

@@ -350,7 +350,7 @@ case class RangeExec(range: org.apache.spark.sql.catalyst.plans.logical.Range)
   override lazy val metrics = Map(
     "numOutputRows" -> SQLMetrics.createMetric(sparkContext, "number of output rows"))
 
-  override lazy val canonicalized: SparkPlan = {
+  override def doCanonicalize(): SparkPlan = {
     RangeExec(range.canonicalized.asInstanceOf[org.apache.spark.sql.catalyst.plans.logical.Range])
   }
 

sql/core/src/main/scala/org/apache/spark/sql/execution/command/cache.scala

@@ -66,10 +66,13 @@ case class UncacheTableCommand(
 /**
  * Clear all cached data from the in-memory cache.
  */
-case object ClearCacheCommand extends RunnableCommand {
+case class ClearCacheCommand() extends RunnableCommand {
 
   override def run(sparkSession: SparkSession): Seq[Row] = {
     sparkSession.catalog.clearCache()
     Seq.empty[Row]
   }
+
+  /** [[org.apache.spark.sql.catalyst.trees.TreeNode.makeCopy()]] does not support 0-arg ctor. */
+  override def makeCopy(newArgs: Array[AnyRef]): ClearCacheCommand = ClearCacheCommand()
 }

sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/LogicalRelation.scala

@@ -35,7 +35,7 @@ case class LogicalRelation(
   extends LeafNode with MultiInstanceRelation {
 
   // Only care about relation when canonicalizing.
-  override lazy val canonicalized: LogicalPlan = copy(
+  override def doCanonicalize(): LogicalPlan = copy(
     output = output.map(QueryPlan.normalizeExprId(_, output)),
     catalogTable = None)