diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala index f41e43431ac1d5072bde9f1fba6acd27eb59d75c..25584de0a92307e3d5c248ab506b1f916e790020 100644 --- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala +++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala @@ -727,7 +727,7 @@ class Analyzer( case p if !p.childrenResolved => p // Replace the index with the related attribute for ORDER BY, // which is a 1-base position of the projection list. - case s @ Sort(orders, global, child) + case Sort(orders, global, child) if orders.exists(_.child.isInstanceOf[UnresolvedOrdinal]) => val newOrders = orders map { case s @ SortOrder(UnresolvedOrdinal(index), direction, nullOrdering) => @@ -744,17 +744,11 @@ class Analyzer( // Replace the index with the corresponding expression in aggregateExpressions. The index is // a 1-base position of aggregateExpressions, which is output columns (select expression) - case a @ Aggregate(groups, aggs, child) if aggs.forall(_.resolved) && + case Aggregate(groups, aggs, child) if aggs.forall(_.resolved) && groups.exists(_.isInstanceOf[UnresolvedOrdinal]) => val newGroups = groups.map { - case ordinal @ UnresolvedOrdinal(index) if index > 0 && index <= aggs.size => - aggs(index - 1) match { - case e if ResolveAggregateFunctions.containsAggregate(e) => - ordinal.failAnalysis( - s"GROUP BY position $index is an aggregate function, and " + - "aggregate functions are not allowed in GROUP BY") - case o => o - } + case u @ UnresolvedOrdinal(index) if index > 0 && index <= aggs.size => + aggs(index - 1) case ordinal @ UnresolvedOrdinal(index) => ordinal.failAnalysis( s"GROUP BY position $index is not in select list " + diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/CheckAnalysis.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/CheckAnalysis.scala index f7109f42838e94b1a5ff25d9352a55773a4b9c54..06bbd39ed11d67c26f15fd2b5aa7cd25685e2113 100644 --- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/CheckAnalysis.scala +++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/CheckAnalysis.scala @@ -266,6 +266,11 @@ trait CheckAnalysis extends PredicateHelper { } def checkValidGroupingExprs(expr: Expression): Unit = { + if (expr.find(_.isInstanceOf[AggregateExpression]).isDefined) { + failAnalysis( + "aggregate functions are not allowed in GROUP BY, but found " + expr.sql) + } + // Check if the data type of expr is orderable. if (!RowOrdering.isOrderable(expr.dataType)) { failAnalysis( @@ -283,8 +288,8 @@ trait CheckAnalysis extends PredicateHelper { } } - aggregateExprs.foreach(checkValidAggregateExpression) groupingExprs.foreach(checkValidGroupingExprs) + aggregateExprs.foreach(checkValidAggregateExpression) case Sort(orders, _, _) => orders.foreach { order => diff --git a/sql/core/src/test/resources/sql-tests/results/group-by-ordinal.sql.out b/sql/core/src/test/resources/sql-tests/results/group-by-ordinal.sql.out index c64520ff93c83653e607a8a5a3a47b6a72c744c9..614c8784ada36a5de190919977d85c572835e5c9 100644 --- a/sql/core/src/test/resources/sql-tests/results/group-by-ordinal.sql.out +++ b/sql/core/src/test/resources/sql-tests/results/group-by-ordinal.sql.out @@ -122,7 +122,7 @@ select a, b, sum(b) from data group by 3 struct<> -- !query 11 output org.apache.spark.sql.AnalysisException -GROUP BY position 3 is an aggregate function, and aggregate functions are not allowed in GROUP BY; line 1 pos 39 +aggregate functions are not allowed in GROUP BY, but found sum(CAST(data.`b` AS BIGINT)); -- !query 12 @@ -131,7 +131,7 @@ select a, b, sum(b) + 2 from data group by 3 struct<> -- !query 12 output org.apache.spark.sql.AnalysisException -GROUP BY position 3 is an aggregate function, and aggregate functions are not allowed in GROUP BY; line 1 pos 43 +aggregate functions are not allowed in GROUP BY, but found (sum(CAST(data.`b` AS BIGINT)) + CAST(2 AS BIGINT)); -- !query 13 diff --git a/sql/core/src/test/scala/org/apache/spark/sql/DataFrameAggregateSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/DataFrameAggregateSuite.scala index e7079120bb7dfe463863450729cde7c14d136231..8569c2d76b694d56273565268524bd07131b87a8 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/DataFrameAggregateSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/DataFrameAggregateSuite.scala @@ -538,4 +538,11 @@ class DataFrameAggregateSuite extends QueryTest with SharedSQLContext { Seq(Row(3, 0, 0.0, 1, 5.0), Row(2, 1, 4.0, 0, 0.0)) ) } + + test("aggregate function in GROUP BY") { + val e = intercept[AnalysisException] { + testData.groupBy(sum($"key")).count() + } + assert(e.message.contains("aggregate functions are not allowed in GROUP BY")) + } }