SPARK-XXXXX ValidateExternalType should return child in error

Author: Mark Andreev

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/objects/objects.scala

@@ -2057,14 +2057,14 @@ case class ValidateExternalType(child: Expression, expected: DataType, externalD
     if (checkType(input)) {
       input
     } else {
-      throw new RuntimeException(s"${input.getClass.getName}$errMsg")
+      throw new RuntimeException(s"${input.getClass.getName}$errMsg at $child")
     }
   }
 
   override def doGenCode(ctx: CodegenContext, ev: ExprCode): ExprCode = {
     // Use unnamed reference that doesn't create a local field here to reduce the number of fields
     // because errMsgField is used only when the type doesn't match.
-    val errMsgField = ctx.addReferenceObj("errMsg", errMsg)
+    val errMsgField = ctx.addReferenceObj("errMsg", errMsg + " at " + child)
     val input = child.genCode(ctx)
     val obj = input.value
     def genCheckTypes(classes: Seq[Class[_]]): String = {

sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/ObjectExpressionsSuite.scala

@@ -556,6 +556,42 @@ class ObjectExpressionsSuite extends SparkFunSuite with ExpressionEvalHelper {
       "java.lang.Integer is not a valid external type for schema of double")
   }
 
+  test("SPARK-XXXXX ValidateExternalType should return child in error") {
+    val inputObject = BoundReference(0, ObjectType(classOf[Row]), nullable = true)
+    Seq(
+      (true, BooleanType),
+      (2.toByte, ByteType),
+      (5.toShort, ShortType),
+      (23, IntegerType),
+      (61L, LongType),
+      (1.0f, FloatType),
+      (10.0, DoubleType),
+      ("abcd".getBytes, BinaryType),
+      ("abcd", StringType),
+      (BigDecimal.valueOf(10), DecimalType.IntDecimal),
+      (IntervalUtils.stringToInterval(UTF8String.fromString("interval 3 day")),
+        CalendarIntervalType),
+      (java.math.BigDecimal.valueOf(10), DecimalType.BigIntDecimal),
+      (Array(3, 2, 1), ArrayType(IntegerType))
+    ).foreach { case (input, dt) =>
+      val enc = RowEncoder.encoderForDataType(dt, lenient = false)
+      val validateType = ValidateExternalType(
+        GetExternalRowField(inputObject, index = 0, fieldName = "c0"),
+        dt,
+        EncoderUtils.lenientExternalDataTypeFor(enc))
+      checkObjectExprEvaluation(validateType, input, InternalRow.fromSeq(Seq(Row(input))))
+    }
+
+    checkExceptionInExpression[RuntimeException](
+      ValidateExternalType(
+        GetExternalRowField(inputObject, index = 0, fieldName = "c0"),
+        DoubleType,
+        DoubleType),
+      InternalRow.fromSeq(Seq(Row(1))),
+      "java.lang.Integer is not a valid external type for schema of double" +
+        " at getexternalrowfield(input[0, org.apache.spark.sql.Row, true], 0, c0)")
+  }
+
   private def javaMapSerializerFor(
       keyClazz: Class[_],
       valueClazz: Class[_])(inputObject: Expression): Expression = {