Github user hequn8128 commented on a diff in the pull request:
https://github.com/apache/flink/pull/5555#discussion_r170577731
--- Diff:
flink-libraries/flink-table/src/main/scala/org/apache/flink/table/runtime/aggregate/AggregateUtil.scala
---
@@ -1393,6 +1393,21 @@ object AggregateUtil {
throw new TableException(s"unsupported Function:
'${unSupported.getName}'")
}
}
+
+ // create distinct accumulator delegate
+ if (aggregateCall.isDistinct) {
--- End diff --
It is better to move this logic into the upper match. Distinct is also a
kind of aggregate whose fields should not be empty. Also, we can reuse some
variables. What do you think?
---
