From 3b7fb84cf88bcae56713fd56396db537fa18f2e5 Mon Sep 17 00:00:00 2001
From: gatorsmile <gatorsmile@gmail.com>
Date: Mon, 13 Jun 2016 13:22:46 -0700
Subject: [PATCH] [SPARK-15676][SQL] Disallow Column Names as Partition Columns
 For Hive Tables

#### What changes were proposed in this pull request?
When creating a Hive Table (not data source tables), a common error users might make is to specify an existing column name as a partition column. Below is what Hive returns in this case:
```
hive> CREATE TABLE partitioned (id bigint, data string) PARTITIONED BY (data string, part string);
FAILED: SemanticException [Error 10035]: Column repeated in partitioning columns
```
Currently, the error we issued is very confusing:
```
org.apache.spark.sql.AnalysisException: org.apache.hadoop.hive.ql.metadata.HiveException: MetaException(message:For direct MetaStore DB connections, we don't support retries at the client level.);
```
This PR is to fix the above issue by capturing the usage error in `Parser`.

#### How was this patch tested?
Added a test case to `DDLCommandSuite`

Author: gatorsmile <gatorsmile@gmail.com>

Closes #13415 from gatorsmile/partitionColumnsInTableSchema.
---
 .../spark/sql/execution/SparkSqlParser.scala    | 17 +++++++++++++++++
 .../sql/execution/command/DDLCommandSuite.scala | 14 ++++++++++++++
 2 files changed, 31 insertions(+)

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 06d8f15dc3..a0508ad601 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 aec7e99d9d..5bee28b446 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(
-- 
GitLab