[SPARK-18125][SQL] Fix a compilation error in codegen due to splitExpression

## What changes were proposed in this pull request?

As reported in the jira, sometimes the generated java code in codegen will cause compilation error.

Code snippet to test it:

    case class Route(src: String, dest: String, cost: Int)
    case class GroupedRoutes(src: String, dest: String, routes: Seq[Route])

    val ds = sc.parallelize(Array(
      Route("a", "b", 1),
      Route("a", "b", 2),
      Route("a", "c", 2),
      Route("a", "d", 10),
      Route("b", "a", 1),
      Route("b", "a", 5),
      Route("b", "c", 6))
    ).toDF.as[Route]

    val grped = ds.map(r => GroupedRoutes(r.src, r.dest, Seq(r)))
      .groupByKey(r => (r.src, r.dest))
      .reduceGroups { (g1: GroupedRoutes, g2: GroupedRoutes) =>
        GroupedRoutes(g1.src, g1.dest, g1.routes ++ g2.routes)
      }.map(_._2)

The problem here is, in `ReferenceToExpressions` we evaluate the children vars to local variables. Then the result expression is evaluated to use those children variables. In the above case, the result expression code is too long and will be split by `CodegenContext.splitExpression`. So those local variables cannot be accessed and cause compilation error.

## How was this patch tested?

Jenkins tests.

Please review https://cwiki.apache.org/confluence/display/SPARK/Contributing+to+Spark before opening a pull request.

Author: Liang-Chi Hsieh <viirya@gmail.com>

Closes #15693 from viirya/fix-codege-compilation-error.

2 files changed, 58 insertions, 6 deletions

diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/ReferenceToExpressions.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/ReferenceToExpressions.scala
index 127797c097..6c75a7a502 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/ReferenceToExpressions.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/ReferenceToExpressions.scala
@@ -63,15 +63,30 @@ case class ReferenceToExpressions(result: Expression, children: Seq[Expression])
 
   override protected def doGenCode(ctx: CodegenContext, ev: ExprCode): ExprCode = {
     val childrenGen = children.map(_.genCode(ctx))
-    val childrenVars = childrenGen.zip(children).map {
-      case (childGen, child) => LambdaVariable(childGen.value, childGen.isNull, child.dataType)
-    }
+    val (classChildrenVars, initClassChildrenVars) = childrenGen.zip(children).map {
+      case (childGen, child) =>
+        // SPARK-18125: The children vars are local variables. If the result expression uses
+        // splitExpression, those variables cannot be accessed so compilation fails.
+        // To fix it, we use class variables to hold those local variables.
+        val classChildVarName = ctx.freshName("classChildVar")
+        val classChildVarIsNull = ctx.freshName("classChildVarIsNull")
+        ctx.addMutableState(ctx.javaType(child.dataType), classChildVarName, "")
+        ctx.addMutableState("boolean", classChildVarIsNull, "")
+
+        val classChildVar =
+          LambdaVariable(classChildVarName, classChildVarIsNull, child.dataType)
+
+        val initCode = s"${classChildVar.value} = ${childGen.value};\n" +
+          s"${classChildVar.isNull} = ${childGen.isNull};"
+
+        (classChildVar, initCode)
+    }.unzip
 
     val resultGen = result.transform {
-      case b: BoundReference => childrenVars(b.ordinal)
+      case b: BoundReference => classChildrenVars(b.ordinal)
     }.genCode(ctx)
 
-    ExprCode(code = childrenGen.map(_.code).mkString("\n") + "\n" + resultGen.code,
-      isNull = resultGen.isNull, value = resultGen.value)
+    ExprCode(code = childrenGen.map(_.code).mkString("\n") + initClassChildrenVars.mkString("\n") +
+      resultGen.code, isNull = resultGen.isNull, value = resultGen.value)
   }
 }
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/DatasetSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/DatasetSuite.scala
index 6fa7b04877..a8dd422aa0 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/DatasetSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/DatasetSuite.scala
@@ -923,6 +923,40 @@ class DatasetSuite extends QueryTest with SharedSQLContext {
         .groupByKey(_.a).flatMapGroups { case (x, iter) => List[Int]() })
   }
 
+  test("SPARK-18125: Spark generated code causes CompileException") {
+    val data = Array(
+      Route("a", "b", 1),
+      Route("a", "b", 2),
+      Route("a", "c", 2),
+      Route("a", "d", 10),
+      Route("b", "a", 1),
+      Route("b", "a", 5),
+      Route("b", "c", 6))
+    val ds = sparkContext.parallelize(data).toDF.as[Route]
+
+    val grped = ds.map(r => GroupedRoutes(r.src, r.dest, Seq(r)))
+      .groupByKey(r => (r.src, r.dest))
+      .reduceGroups { (g1: GroupedRoutes, g2: GroupedRoutes) =>
+        GroupedRoutes(g1.src, g1.dest, g1.routes ++ g2.routes)
+      }.map(_._2)
+
+    val expected = Seq(
+      GroupedRoutes("a", "d", Seq(Route("a", "d", 10))),
+      GroupedRoutes("b", "c", Seq(Route("b", "c", 6))),
+      GroupedRoutes("a", "b", Seq(Route("a", "b", 1), Route("a", "b", 2))),
+      GroupedRoutes("b", "a", Seq(Route("b", "a", 1), Route("b", "a", 5))),
+      GroupedRoutes("a", "c", Seq(Route("a", "c", 2)))
+    )
+
+    implicit def ordering[GroupedRoutes]: Ordering[GroupedRoutes] = new Ordering[GroupedRoutes] {
+      override def compare(x: GroupedRoutes, y: GroupedRoutes): Int = {
+        x.toString.compareTo(y.toString)
+      }
+    }
+
+    checkDatasetUnorderly(grped, expected: _*)
+  }
+
   test("SPARK-18189: Fix serialization issue in KeyValueGroupedDataset") {
     val resultValue = 12345
     val keyValueGrouped = Seq((1, 2), (3, 4)).toDS().groupByKey(_._1)
@@ -1071,3 +1105,6 @@ object DatasetTransform {
     ds.map(_ + 1)
   }
 }
+
+case class Route(src: String, dest: String, cost: Int)
+case class GroupedRoutes(src: String, dest: String, routes: Seq[Route])