diff --git a/sql/core/src/main/scala/org/apache/spark/sql/execution/basicOperators.scala b/sql/core/src/main/scala/org/apache/spark/sql/execution/basicOperators.scala index 247c900baae9e2bd4a1551880ed00c32fc720f21..77b98064a9e161f3194e912bb586b6f3fc52787e 100644 --- a/sql/core/src/main/scala/org/apache/spark/sql/execution/basicOperators.scala +++ b/sql/core/src/main/scala/org/apache/spark/sql/execution/basicOperators.scala @@ -237,7 +237,10 @@ case class TakeOrderedAndProject( projectList: Option[Seq[NamedExpression]], child: SparkPlan) extends UnaryNode { - override def output: Seq[Attribute] = child.output + override def output: Seq[Attribute] = { + val projectOutput = projectList.map(_.map(_.toAttribute)) + projectOutput.getOrElse(child.output) + } override def outputPartitioning: Partitioning = SinglePartition @@ -263,6 +266,13 @@ case class TakeOrderedAndProject( protected override def doExecute(): RDD[InternalRow] = sparkContext.makeRDD(collectData(), 1) override def outputOrdering: Seq[SortOrder] = sortOrder + + override def simpleString: String = { + val orderByString = sortOrder.mkString("[", ",", "]") + val outputString = output.mkString("[", ",", "]") + + s"TakeOrderedAndProject(limit=$limit, orderBy=$orderByString, output=$outputString)" + } } /** diff --git a/sql/core/src/test/scala/org/apache/spark/sql/execution/PlannerSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/execution/PlannerSuite.scala index 937a108543531918b320481a127ad52568fab7c1..fad93b014c2378e7641858f31bd5c6868566b117 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/execution/PlannerSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/execution/PlannerSuite.scala @@ -162,9 +162,23 @@ class PlannerSuite extends SparkFunSuite with SharedSQLContext { } test("efficient limit -> project -> sort") { - val query = testData.sort('key).select('value).limit(2).logicalPlan - val planned = ctx.planner.TakeOrderedAndProject(query) - assert(planned.head.isInstanceOf[execution.TakeOrderedAndProject]) + { + val query = + testData.select('key, 'value).sort('key).limit(2).logicalPlan + val planned = ctx.planner.TakeOrderedAndProject(query) + assert(planned.head.isInstanceOf[execution.TakeOrderedAndProject]) + assert(planned.head.output === testData.select('key, 'value).logicalPlan.output) + } + + { + // We need to make sure TakeOrderedAndProject's output is correct when we push a project + // into it. + val query = + testData.select('key, 'value).sort('key).select('value, 'key).limit(2).logicalPlan + val planned = ctx.planner.TakeOrderedAndProject(query) + assert(planned.head.isInstanceOf[execution.TakeOrderedAndProject]) + assert(planned.head.output === testData.select('value, 'key).logicalPlan.output) + } } test("PartitioningCollection") {