Github user twalthr commented on a diff in the pull request:
https://github.com/apache/flink/pull/5040#discussion_r152614003
--- Diff:
flink-libraries/flink-table/src/main/scala/org/apache/flink/table/calcite/FlinkTypeFactory.scala
---
@@ -417,6 +417,10 @@ object FlinkTypeFactory {
val compositeRelDataType =
relDataType.asInstanceOf[CompositeRelDataType]
compositeRelDataType.compositeType
+ case ROW if relDataType.isInstanceOf[RelRecordType] =>
+ val relRecordType = relDataType.asInstanceOf[RelRecordType]
+ new RowSchema(relRecordType).typeInfo
+
// ROW and CURSOR for UDTF case, whose type info will never be used,
just a placeholder
case ROW | CURSOR => new NothingTypeInfo
--- End diff --
Maybe we can remove the `ROW` here if it has no impact on the existing
tests.
---