[SPARK-10449] [SQL] Don't merge decimal types with incompatable precision or scales

From JIRA: Schema merging should only handle struct fields. But currently we also reconcile decimal precision and scale information.

Author: Holden Karau <holden@pigscanfly.ca>

Closes #8634 from holdenk/SPARK-10449-dont-merge-different-precision.

sql/catalyst/src/main/scala/org/apache/spark/sql/types/StructType.scala

@@ -373,10 +373,19 @@ object StructType extends AbstractDataType {
         StructType(newFields)
 
       case (DecimalType.Fixed(leftPrecision, leftScale),
-      DecimalType.Fixed(rightPrecision, rightScale)) =>
-        DecimalType(
-          max(leftScale, rightScale) + max(leftPrecision - leftScale, rightPrecision - rightScale),
-          max(leftScale, rightScale))
+        DecimalType.Fixed(rightPrecision, rightScale)) =>
+        if ((leftPrecision == rightPrecision) && (leftScale == rightScale)) {
+          DecimalType(leftPrecision, leftScale)
+        } else if ((leftPrecision != rightPrecision) && (leftScale != rightScale)) {
+          throw new SparkException("Failed to merge Decimal Tpes with incompatible " +
+            s"precision $leftPrecision and $rightPrecision & scale $leftScale and $rightScale")
+        } else if (leftPrecision != rightPrecision) {
+          throw new SparkException("Failed to merge Decimal Tpes with incompatible " +
+            s"precision $leftPrecision and $rightPrecision")
+        } else {
+          throw new SparkException("Failed to merge Decimal Tpes with incompatible " +
+            s"scala $leftScale and $rightScale")
+        }
 
       case (leftUdt: UserDefinedType[_], rightUdt: UserDefinedType[_])
         if leftUdt.userClass == rightUdt.userClass => leftUdt