[SPARK-35069][SQL] TRANSFORM forbidden DISTICNT and ALL, also make the error clear

sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBase.g4

@@ -524,9 +524,9 @@ querySpecification
     ;
 
 transformClause
-    : (SELECT kind=TRANSFORM '(' namedExpressionSeq ')'
-            | kind=MAP namedExpressionSeq
-            | kind=REDUCE namedExpressionSeq)
+    : (SELECT kind=TRANSFORM '(' setQuantifier? namedExpressionSeq ')'
+            | kind=MAP setQuantifier? namedExpressionSeq
+            | kind=REDUCE setQuantifier? namedExpressionSeq)
       inRowFormat=rowFormat?
       (RECORDWRITER recordWriter=STRING)?
       USING script=STRING

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala

@@ -660,6 +660,9 @@ class AstBuilder extends SqlBaseBaseVisitor[AnyRef] with SQLConfHelper with Logg
       havingClause: HavingClauseContext,
       windowClause: WindowClauseContext,
       relation: LogicalPlan): LogicalPlan = withOrigin(ctx) {
+    if (transformClause.setQuantifier != null) {
+      throw QueryParsingErrors.transformNotSupportQuantifierError(transformClause.setQuantifier)
+    }
     // Create the attributes.
     val (attributes, schemaLess) = if (transformClause.colTypeList != null) {
       // Typed return columns.

sql/catalyst/src/main/scala/org/apache/spark/sql/errors/QueryParsingErrors.scala

@@ -88,6 +88,11 @@ object QueryParsingErrors {
     new ParseException("DISTRIBUTE BY is not supported", ctx)
   }
 
+  def transformNotSupportQuantifierError(ctx: ParserRuleContext): Throwable = {
+    new ParseException("TRANSFORM not support quantifier DISTINCT/ALL " +
+      "for input expression seq", ctx)
+  }
+
   def transformWithSerdeUnsupportedError(ctx: ParserRuleContext): Throwable = {
     new ParseException("TRANSFORM with serde is only supported in hive mode", ctx)
   }

sql/core/src/test/resources/sql-tests/inputs/transform.sql

@@ -311,7 +311,7 @@ SET spark.sql.legacy.parser.havingWithoutGroupByAsWhere=false;
 SET spark.sql.parser.quotedRegexColumnNames=true;
 
 SELECT TRANSFORM(`(a|b)?+.+`)
- USING 'cat' AS (c)
+  USING 'cat' AS (c)
 FROM script_trans;
 
 SET spark.sql.parser.quotedRegexColumnNames=false;
@@ -320,4 +320,14 @@ SET spark.sql.parser.quotedRegexColumnNames=false;
 WITH temp AS (
   SELECT TRANSFORM(a) USING 'cat' AS (b string) FROM t
 )
-SELECT t1.b FROM temp t1 JOIN temp t2 ON t1.b = t2.b
+SELECT t1.b FROM temp t1 JOIN temp t2 ON t1.b = t2.b;
+
+SELECT TRANSFORM(DISTINCT b, a, c)
+  USING 'cat' AS (a, b, c)
+FROM script_trans
+WHERE a <= 4;
+
+SELECT TRANSFORM(ALL b, a, c)
+  USING 'cat' AS (a, b, c)
+FROM script_trans
+WHERE a <= 4;

sql/core/src/test/resources/sql-tests/results/transform.sql.out

@@ -1,5 +1,5 @@
 -- Automatically generated by SQLQueryTestSuite
--- Number of queries: 42
+-- Number of queries: 44
 
 
 -- !query
@@ -610,7 +610,7 @@ spark.sql.parser.quotedRegexColumnNames	true
 
 -- !query
 SELECT TRANSFORM(`(a|b)?+.+`)
- USING 'cat' AS (c)
+  USING 'cat' AS (c)
 FROM script_trans
 -- !query schema
 struct<c:string>
@@ -627,6 +627,7 @@ struct<key:string,value:string>
 -- !query output
 spark.sql.parser.quotedRegexColumnNames	false
 
+
 -- !query
 WITH temp AS (
   SELECT TRANSFORM(a) USING 'cat' AS (b string) FROM t
@@ -638,3 +639,43 @@ struct<b:string>
 1
 2
 3
+
+
+-- !query
+SELECT TRANSFORM(DISTINCT b, a, c)
+  USING 'cat' AS (a, b, c)
+FROM script_trans
+WHERE a <= 4
+-- !query schema
+struct<>
+-- !query output
+org.apache.spark.sql.catalyst.parser.ParseException
+
+TRANSFORM not support quantifier DISTINCT/ALL for input expression seq(line 1, pos 17)
+
+== SQL ==
+SELECT TRANSFORM(DISTINCT b, a, c)
+-----------------^^^
+  USING 'cat' AS (a, b, c)
+FROM script_trans
+WHERE a <= 4
+
+
+-- !query
+SELECT TRANSFORM(ALL b, a, c)
+  USING 'cat' AS (a, b, c)
+FROM script_trans
+WHERE a <= 4
+-- !query schema
+struct<>
+-- !query output
+org.apache.spark.sql.catalyst.parser.ParseException
+
+TRANSFORM not support quantifier DISTINCT/ALL for input expression seq(line 1, pos 17)
+
+== SQL ==
+SELECT TRANSFORM(ALL b, a, c)
+-----------------^^^
+  USING 'cat' AS (a, b, c)
+FROM script_trans
+WHERE a <= 4