diff --git a/sql/core/src/main/scala/org/apache/spark/sql/execution/SparkSqlParser.scala b/sql/core/src/main/scala/org/apache/spark/sql/execution/SparkSqlParser.scala index 06d8f15dc300825fafb850702bd863ba2b856047..a0508ad6019bdf6ccde020bca6df5e44bb1eef8d 100644 --- a/sql/core/src/main/scala/org/apache/spark/sql/execution/SparkSqlParser.scala +++ b/sql/core/src/main/scala/org/apache/spark/sql/execution/SparkSqlParser.scala @@ -903,6 +903,23 @@ class SparkSqlAstBuilder(conf: SQLConf) extends AstBuilder { val properties = Option(ctx.tablePropertyList).map(visitPropertyKeyValues).getOrElse(Map.empty) val selectQuery = Option(ctx.query).map(plan) + // Ensuring whether no duplicate name is used in table definition + val colNames = cols.map(_.name) + if (colNames.length != colNames.distinct.length) { + val duplicateColumns = colNames.groupBy(identity).collect { + case (x, ys) if ys.length > 1 => "\"" + x + "\"" + } + throw operationNotAllowed(s"Duplicated column names found in table definition of $name: " + + duplicateColumns.mkString("[", ",", "]"), ctx) + } + + // For Hive tables, partition columns must not be part of the schema + val badPartCols = partitionCols.map(_.name).toSet.intersect(colNames.toSet) + if (badPartCols.nonEmpty) { + throw operationNotAllowed(s"Partition columns may not be specified in the schema: " + + badPartCols.map("\"" + _ + "\"").mkString("[", ",", "]"), ctx) + } + // Note: Hive requires partition columns to be distinct from the schema, so we need // to include the partition columns here explicitly val schema = cols ++ partitionCols diff --git a/sql/core/src/test/scala/org/apache/spark/sql/execution/command/DDLCommandSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/execution/command/DDLCommandSuite.scala index aec7e99d9d37dbb674dc82711aafd8c13f47eca5..5bee28b4462c662ea1cbfff5c5d20d6e4671d94e 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/execution/command/DDLCommandSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/execution/command/DDLCommandSuite.scala @@ -334,6 +334,20 @@ class DDLCommandSuite extends PlanTest { assert(ct.table.storage.locationUri == Some("/something/anything")) } + test("create table - column repeated in partitioning columns") { + val query = "CREATE TABLE tab1 (key INT, value STRING) PARTITIONED BY (key INT, hr STRING)" + val e = intercept[ParseException] { parser.parsePlan(query) } + assert(e.getMessage.contains( + "Operation not allowed: Partition columns may not be specified in the schema: [\"key\"]")) + } + + test("create table - duplicate column names in the table definition") { + val query = "CREATE TABLE default.tab1 (key INT, key STRING)" + val e = intercept[ParseException] { parser.parsePlan(query) } + assert(e.getMessage.contains("Operation not allowed: Duplicated column names found in " + + "table definition of `default`.`tab1`: [\"key\"]")) + } + test("create table using - with partitioned by") { val query = "CREATE TABLE my_tab(a INT, b STRING) USING parquet PARTITIONED BY (a)" val expected = CreateTableUsing(