[SPARK-15825] [SQL] Fix SMJ invalid results

## What changes were proposed in this pull request? Code generated `SortMergeJoin` failed with wrong results when using structs as keys. This could (eventually) be traced back to the use of a wrong row reference when comparing structs. ## How was this patch tested? TBD Author: Herman van Hovell <[email protected]> Closes apache#13589 from hvanhovell/SPARK-15822.
jjallaire · Jun 10, 2016 · e05a2fe · e05a2fe
1 parent 026eb90
commit e05a2fe
Show file tree

Hide file tree

Showing 2 changed files with 16 additions and 0 deletions.
diff --git a/...lyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala b/...lyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala
@@ -490,6 +490,7 @@ class CodegenContext {
       addNewFunction(compareFunc, funcCode)
       s"this.$compareFunc($c1, $c2)"
     case schema: StructType =>
+      INPUT_ROW = "i"
       val comparisons = GenerateOrdering.genComparisons(this, schema)
       val compareFunc = freshName("compareStruct")
       val funcCode: String =

diff --git a/sql/core/src/test/scala/org/apache/spark/sql/execution/joins/InnerJoinSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/execution/joins/InnerJoinSuite.scala
@@ -271,4 +271,19 @@ class InnerJoinSuite extends SparkPlanTest with SharedSQLContext {
       )
     )
   }
+
+  {
+    def df: DataFrame = spark.range(3).selectExpr("struct(id, id) as key", "id as value")
+    lazy val left = df.selectExpr("key", "concat('L', value) as value").alias("left")
+    lazy val right = df.selectExpr("key", "concat('R', value) as value").alias("right")
+    testInnerJoin(
+      "SPARK-15822 - test structs as keys",
+      left,
+      right,
+      () => (left.col("key") === right.col("key")).expr,
+      Seq(
+        (Row(0, 0), "L0", Row(0, 0), "R0"),
+        (Row(1, 1), "L1", Row(1, 1), "R1"),
+        (Row(2, 2), "L2", Row(2, 2), "R2")))
+  }
 }