[SPARK-24373][SQL] Add AnalysisBarrier to RelationalGroupedDataset's …

…and KeyValueGroupedDataset's child

When we create a `RelationalGroupedDataset` or a `KeyValueGroupedDataset` we set its child to the `logicalPlan` of the `DataFrame` we need to aggregate. Since the `logicalPlan` is already analyzed, we should not analyze it again. But this happens when the new plan of the aggregate is analyzed.

The current behavior in most of the cases is likely to produce no harm, but in other cases re-analyzing an analyzed plan can change it, since the analysis is not idempotent. This can cause issues like the one described in the JIRA (missing to find a cached plan).

The PR adds an `AnalysisBarrier` to the `logicalPlan` which is used as child of `RelationalGroupedDataset` or a `KeyValueGroupedDataset`.

added UT

Author: Marco Gaido <marcogaido91@gmail.com>

Closes #21432 from mgaido91/SPARK-24373.

(cherry picked from commit de01a8d)
Signed-off-by: Wenchen Fan <wenchen@databricks.com>

sql/core/src/main/scala/org/apache/spark/sql/Dataset.scala

@@ -196,7 +196,7 @@ class Dataset[T] private[sql](
   }
 
   // Wraps analyzed logical plans with an analysis barrier so we won't traverse/resolve it again.
-  @transient private val planWithBarrier = AnalysisBarrier(logicalPlan)
+  @transient private[sql] val planWithBarrier = AnalysisBarrier(logicalPlan)
 
   /**
    * Currently [[ExpressionEncoder]] is the only implementation of [[Encoder]], here we turn the

sql/core/src/main/scala/org/apache/spark/sql/KeyValueGroupedDataset.scala

@@ -49,7 +49,7 @@ class KeyValueGroupedDataset[K, V] private[sql](
   private implicit val kExprEnc = encoderFor(kEncoder)
   private implicit val vExprEnc = encoderFor(vEncoder)
 
-  private def logicalPlan = queryExecution.analyzed
+  private def logicalPlan = AnalysisBarrier(queryExecution.analyzed)
   private def sparkSession = queryExecution.sparkSession
 
   /**

sql/core/src/main/scala/org/apache/spark/sql/RelationalGroupedDataset.scala

@@ -64,17 +64,17 @@ class RelationalGroupedDataset protected[sql](
     groupType match {
       case RelationalGroupedDataset.GroupByType =>
         Dataset.ofRows(
-          df.sparkSession, Aggregate(groupingExprs, aliasedAgg, df.logicalPlan))
+          df.sparkSession, Aggregate(groupingExprs, aliasedAgg, df.planWithBarrier))
       case RelationalGroupedDataset.RollupType =>
         Dataset.ofRows(
-          df.sparkSession, Aggregate(Seq(Rollup(groupingExprs)), aliasedAgg, df.logicalPlan))
+          df.sparkSession, Aggregate(Seq(Rollup(groupingExprs)), aliasedAgg, df.planWithBarrier))
       case RelationalGroupedDataset.CubeType =>
         Dataset.ofRows(
-          df.sparkSession, Aggregate(Seq(Cube(groupingExprs)), aliasedAgg, df.logicalPlan))
+          df.sparkSession, Aggregate(Seq(Cube(groupingExprs)), aliasedAgg, df.planWithBarrier))
       case RelationalGroupedDataset.PivotType(pivotCol, values) =>
         val aliasedGrps = groupingExprs.map(alias)
         Dataset.ofRows(
-          df.sparkSession, Pivot(aliasedGrps, pivotCol, values, aggExprs, df.logicalPlan))
+          df.sparkSession, Pivot(aliasedGrps, pivotCol, values, aggExprs, df.planWithBarrier))
     }
   }
 
@@ -434,7 +434,7 @@ class RelationalGroupedDataset protected[sql](
           df.exprEnc.schema,
           groupingAttributes,
           df.logicalPlan.output,
-          df.logicalPlan))
+          df.planWithBarrier))
   }
 
   /**
@@ -460,7 +460,7 @@ class RelationalGroupedDataset protected[sql](
       case other => Alias(other, other.toString)()
     }
     val groupingAttributes = groupingNamedExpressions.map(_.toAttribute)
-    val child = df.logicalPlan
+    val child = df.planWithBarrier
     val project = Project(groupingNamedExpressions ++ child.output, child)
     val output = expr.dataType.asInstanceOf[StructType].toAttributes
     val plan = FlatMapGroupsInPandas(groupingAttributes, expr, output, project)

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

@@ -0,0 +1,96 @@
+/*
+ * 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
+
+import org.apache.spark.api.python.PythonEvalType
+import org.apache.spark.sql.catalyst.plans.logical.AnalysisBarrier
+import org.apache.spark.sql.execution.python.PythonUDF
+import org.apache.spark.sql.functions.udf
+import org.apache.spark.sql.test.SharedSQLContext
+import org.apache.spark.sql.types.{LongType, StructField, StructType}
+
+class GroupedDatasetSuite extends QueryTest with SharedSQLContext {
+  import testImplicits._
+
+  private val scalaUDF = udf((x: Long) => { x + 1 })
+  private lazy val datasetWithUDF = spark.range(1).toDF("s").select($"s", scalaUDF($"s"))
+
+  private def assertContainsAnalysisBarrier(ds: Dataset[_], atLevel: Int = 1): Unit = {
+    assert(atLevel >= 0)
+    var children = Seq(ds.queryExecution.logical)
+    (1 to atLevel).foreach { _ =>
+      children = children.flatMap(_.children)
+    }
+    val barriers = children.collect {
+      case ab: AnalysisBarrier => ab
+    }
+    assert(barriers.nonEmpty, s"Plan does not contain AnalysisBarrier at level $atLevel:\n" +
+      ds.queryExecution.logical)
+  }
+
+  test("SPARK-24373: avoid running Analyzer rules twice on RelationalGroupedDataset") {
+    val groupByDataset = datasetWithUDF.groupBy()
+    val rollupDataset = datasetWithUDF.rollup("s")
+    val cubeDataset = datasetWithUDF.cube("s")
+    val pivotDataset = datasetWithUDF.groupBy().pivot("s", Seq(1, 2))
+    datasetWithUDF.cache()
+    Seq(groupByDataset, rollupDataset, cubeDataset, pivotDataset).foreach { rgDS =>
+      val df = rgDS.count()
+      assertContainsAnalysisBarrier(df)
+      assertCached(df)
+    }
+
+    val flatMapGroupsInRDF = datasetWithUDF.groupBy().flatMapGroupsInR(
+      Array.emptyByteArray,
+      Array.emptyByteArray,
+      Array.empty,
+      StructType(Seq(StructField("s", LongType))))
+    val flatMapGroupsInPandasDF = datasetWithUDF.groupBy().flatMapGroupsInPandas(PythonUDF(
+      "pyUDF",
+      null,
+      StructType(Seq(StructField("s", LongType))),
+      Seq.empty,
+      PythonEvalType.SQL_GROUPED_MAP_PANDAS_UDF,
+      true))
+    Seq(flatMapGroupsInRDF, flatMapGroupsInPandasDF).foreach { df =>
+      assertContainsAnalysisBarrier(df, 2)
+      assertCached(df)
+    }
+    datasetWithUDF.unpersist(true)
+  }
+
+  test("SPARK-24373: avoid running Analyzer rules twice on KeyValueGroupedDataset") {
+    val kvDasaset = datasetWithUDF.groupByKey(_.getLong(0))
+    datasetWithUDF.cache()
+    val mapValuesKVDataset = kvDasaset.mapValues(_.getLong(0)).reduceGroups(_ + _)
+    val keysKVDataset = kvDasaset.keys
+    val flatMapGroupsKVDataset = kvDasaset.flatMapGroups((k, _) => Seq(k))
+    val aggKVDataset = kvDasaset.count()
+    val otherKVDataset = spark.range(1).groupByKey(_ + 1)
+    val cogroupKVDataset = kvDasaset.cogroup(otherKVDataset)((k, _, _) => Seq(k))
+    Seq((mapValuesKVDataset, 1),
+        (keysKVDataset, 2),
+        (flatMapGroupsKVDataset, 2),
+        (aggKVDataset, 1),
+        (cogroupKVDataset, 2)).foreach { case (df, analysisBarrierDepth) =>
+      assertContainsAnalysisBarrier(df, analysisBarrierDepth)
+      assertCached(df)
+    }
+    datasetWithUDF.unpersist(true)
+  }
+}