HuangXingBo commented on a change in pull request #10086: [FLINK-14584][python]
Support complex data types in Python user-defined functions
URL: https://github.com/apache/flink/pull/10086#discussion_r354773824
##########
File path:
flink-python/src/main/java/org/apache/flink/table/runtime/typeutils/PythonTypeUtils.java
##########
@@ -131,6 +187,29 @@ public TypeSerializer visit(DateType dateType) {
return DateSerializer.INSTANCE;
}
+ @Override
+ @SuppressWarnings("unchecked")
+ public TypeSerializer visit(ArrayType arrayType) {
+ LogicalType elementType = arrayType.getElementType();
+ TypeSerializer<?> elementTypeSerializer =
elementType.accept(this);
+ Class<?> elementClass =
LogicalTypeToConversionClassConverter.INSTANCE.visit(elementType);
Review comment:
As we disscussed offline, logicalType does not contain class information,
so we can't use LogicalTypeToConversionClassConverter to get the correct
conversion class information of DateType, TimeType,TimestampType and ArrayType
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
