Github user twalthr commented on a diff in the pull request:
https://github.com/apache/flink/pull/5212#discussion_r159393961
--- Diff:
flink-libraries/flink-table/src/main/scala/org/apache/flink/table/codegen/calls/ScalarOperators.scala
---
@@ -47,9 +47,22 @@ object ScalarOperators {
nullCheck: Boolean,
resultType: TypeInformation[_],
left: GeneratedExpression,
- right: GeneratedExpression)
- : GeneratedExpression = {
- val leftCasting = numericCasting(left.resultType, resultType)
+ right: GeneratedExpression): GeneratedExpression = {
+
+ val leftCasting = operator match {
+ case "%" =>
+ if (left.resultType == right.resultType) {
+ numericCasting(left.resultType, resultType)
+ } else {
+ val castedType = if (isDecimal(left.resultType)) {
+ Types.LONG
--- End diff --
Wouldn't it be better to use `BigDecimal. remainder` instead of losing
precision?
---
