Remove any Filters with DynamicPruning that didn't filter on partitio…

…n column
apache · Aug 21, 2020 · 98f7275 · 98f7275
1 parent bcc81be
commit 98f7275
Show file tree

Hide file tree

Showing 4 changed files with 24 additions and 3 deletions.
diff --git a/sql/core/src/main/scala/org/apache/spark/sql/execution/SparkOptimizer.scala b/sql/core/src/main/scala/org/apache/spark/sql/execution/SparkOptimizer.scala
@@ -44,10 +44,12 @@ class SparkOptimizer(
     Batch("PartitionPruning", Once,
       PartitionPruning,
       OptimizeSubqueries) :+
-    Batch("Pushdown Filters from PartitionPruning", fixedPoint,
+    Batch("Pushdown Filters from PartitionPruning before Inferring Filters", fixedPoint,
       PushDownPredicates) :+
     Batch("Infer Filters from PartitionPruning", Once,
       InferFiltersFromConstraints) :+
+    Batch("Pushdown Filters from PartitionPruning after Inferring Filters", fixedPoint,
+      PushDownPredicates) :+
     Batch("Cleanup filters that cannot be pushed down", Once,
       CleanupDynamicPruningFilters,
       PruneFilters)) ++

diff --git a/...in/scala/org/apache/spark/sql/execution/dynamicpruning/CleanupDynamicPruningFilters.scala b/...in/scala/org/apache/spark/sql/execution/dynamicpruning/CleanupDynamicPruningFilters.scala
@@ -17,7 +17,7 @@
 
 package org.apache.spark.sql.execution.dynamicpruning
 
-import org.apache.spark.sql.catalyst.expressions.{DynamicPruning, PredicateHelper}
+import org.apache.spark.sql.catalyst.expressions.{DynamicPruning, DynamicPruningSubquery, Expression, PredicateHelper}
 import org.apache.spark.sql.catalyst.expressions.Literal.TrueLiteral
 import org.apache.spark.sql.catalyst.planning.PhysicalOperation
 import org.apache.spark.sql.catalyst.plans.logical.{Filter, LogicalPlan}
@@ -32,12 +32,29 @@ import org.apache.spark.sql.internal.SQLConf
  */
 object CleanupDynamicPruningFilters extends Rule[LogicalPlan] with PredicateHelper {
 
+  private def isFilterOnNonPartition(condition: Expression, child: LogicalPlan): Boolean = {
+    splitConjunctivePredicates(condition).exists {
+      case DynamicPruningSubquery(pruningKey, _, _, _, _, _) =>
+        PartitionPruning.getPartitionTableScan(pruningKey, child).isEmpty
+      case _ => false
+    }
+  }
+
   override def apply(plan: LogicalPlan): LogicalPlan = {
     if (!SQLConf.get.dynamicPartitionPruningEnabled) {
       return plan
     }
 
     plan.transform {
+      // Remove any Filters with DynamicPruning that didn't filter on partition column`.
+      // This is inferred by Infer Filters from PartitionPruning.
+      case f @ Filter(condition, child) if isFilterOnNonPartition(condition, child) =>
+        val newCondition = condition.transform {
+          case DynamicPruningSubquery(pruningKey, _, _, _, _, _)
+            if PartitionPruning.getPartitionTableScan(pruningKey, child).isEmpty =>
+            TrueLiteral
+        }
+        f.copy(condition = newCondition)
       // pass through anything that is pushed down into PhysicalOperation
       case p @ PhysicalOperation(_, _, LogicalRelation(_: HadoopFsRelation, _, _, _)) => p
       // remove any Filters with DynamicPruning that didn't get pushed down to PhysicalOperation.

diff --git a/sql/core/src/main/scala/org/apache/spark/sql/execution/dynamicpruning/PartitionPruning.scala b/sql/core/src/main/scala/org/apache/spark/sql/execution/dynamicpruning/PartitionPruning.scala
@@ -147,7 +147,7 @@ object PartitionPruning extends Rule[LogicalPlan] with PredicateHelper {
     }
 
     // the pruning overhead is the total size in bytes of all scan relations
-    val overhead = otherPlan.collectLeaves().map(_.stats.sizeInBytes).sum.toFloat
+    val overhead = otherPlan.collectLeaves().map(_.stats.sizeInBytes).sum
 
     filterRatio * partPlan.stats.sizeInBytes.toFloat > overhead.toFloat
   }

diff --git a/sql/core/src/test/scala/org/apache/spark/sql/DynamicPartitionPruningSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/DynamicPartitionPruningSuite.scala
@@ -1299,6 +1299,7 @@ abstract class DynamicPartitionPruningSuiteBase
     assert(collectDynamicPruningExpressions(df.queryExecution.executedPlan).size === 2)
     checkDistinctSubqueries(df, 1)
     checkPartitionPruningPredicate(df, false, true)
+    assert(!checkUnpushedFilters(df), "Inferred dynamic pruning expression has been pushed down.")
 
     checkAnswer(df,
       Row(2, 20, 2) ::
@@ -1324,6 +1325,7 @@ abstract class DynamicPartitionPruningSuiteBase
 
     assert(collectDynamicPruningExpressions(df.queryExecution.executedPlan).size === 1)
     checkPartitionPruningPredicate(df, false, true)
+    assert(!checkUnpushedFilters(df), "Inferred dynamic pruning expression should be removed.")
 
     checkAnswer(df,
       Row(2, "NL", 2) ::