[SPARK-2965][SQL] Fix HashOuterJoin output nullabilities.

Output attributes of opposite side of `OuterJoin` should be nullable.

Author: Takuya UESHIN <ueshin@happy-camper.st>

Closes #1887 from ueshin/issues/SPARK-2965 and squashes the following commits:

bcb2d37 [Takuya UESHIN] Fix HashOuterJoin output nullabilities.

1 files changed, 12 insertions, 1 deletions

diff --git a/sql/core/src/main/scala/org/apache/spark/sql/execution/joins.scala b/sql/core/src/main/scala/org/apache/spark/sql/execution/joins.scala
index 51bb615307..ea075f8c65 100644
--- a/sql/core/src/main/scala/org/apache/spark/sql/execution/joins.scala
+++ b/sql/core/src/main/scala/org/apache/spark/sql/execution/joins.scala
@@ -168,7 +168,18 @@ case class HashOuterJoin(
   override def requiredChildDistribution =
     ClusteredDistribution(leftKeys) :: ClusteredDistribution(rightKeys) :: Nil
 
-  def output = left.output ++ right.output
+  override def output = {
+    joinType match {
+      case LeftOuter =>
+        left.output ++ right.output.map(_.withNullability(true))
+      case RightOuter =>
+        left.output.map(_.withNullability(true)) ++ right.output
+      case FullOuter =>
+        left.output.map(_.withNullability(true)) ++ right.output.map(_.withNullability(true))
+      case x =>
+        throw new Exception(s"HashOuterJoin should not take $x as the JoinType")
+    }
+  }
 
   // TODO we need to rewrite all of the iterators with our own implementation instead of the Scala
   // iterator for performance purpose.