[SPARK-4650][SQL] Supporting multi column support in countDistinct function...

[SPARK-4650][SQL] Supporting multi column support in countDistinct function like count(distinct c1,c2..) in Spark SQL

Supporting multi column support in countDistinct function like count(distinct c1,c2..) in Spark SQL

Author: ravipesala <ravindra.pesala@huawei.com>
Author: Michael Armbrust <michael@databricks.com>

Closes #3511 from ravipesala/countdistinct and squashes the following commits:

cc4dbb1 [ravipesala] style
070e12a [ravipesala] Supporting multi column support in count(distinct c1,c2..) in Spark SQL

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/SqlParser.scala

@@ -277,7 +277,8 @@ class SqlParser extends AbstractSparkSQLParser {
     | SUM   ~> "(" ~> DISTINCT ~> expression <~ ")" ^^ { case exp => SumDistinct(exp) }
     | COUNT ~  "(" ~> "*"                    <~ ")" ^^ { case _ => Count(Literal(1)) }
     | COUNT ~  "(" ~> expression             <~ ")" ^^ { case exp => Count(exp) }
-    | COUNT ~> "(" ~> DISTINCT ~> expression <~ ")" ^^ { case exp => CountDistinct(exp :: Nil) }
+    | COUNT ~> "(" ~> DISTINCT ~> repsep(expression, ",") <~ ")" ^^
+      { case exps => CountDistinct(exps) }
     | APPROXIMATE ~ COUNT ~ "(" ~ DISTINCT ~> expression <~ ")" ^^
       { case exp => ApproxCountDistinct(exp) }
     | APPROXIMATE ~> "(" ~> floatLit ~ ")" ~ COUNT ~ "(" ~ DISTINCT ~ expression <~ ")" ^^

sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala

@@ -992,4 +992,11 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll {
       "nulldata2 on nulldata1.value <=> nulldata2.value"),
         (1 to 2).map(i => Seq(i)))
   }
+
+  test("Multi-column COUNT(DISTINCT ...)") {
+    val data = TestData(1,"val_1") :: TestData(2,"val_2") :: Nil
+    val rdd = sparkContext.parallelize((0 to 1).map(i => data(i)))
+    rdd.registerTempTable("distinctData")
+    checkAnswer(sql("SELECT COUNT(DISTINCT key,value) FROM distinctData"), 2)
+  }
 }