Details
-
New Feature
-
Status: Open
-
P3
-
Resolution: Unresolved
-
None
-
None
-
None
Description
The Beam documentation mentions that Flatten does not require coders of all inputs to be of the same type, as long as the result type is the same. However, the current implementation in the Spark Runner requires all coders to match.
Attachments
Issue Links
- is a clone of
-
BEAM-6523 Support transcoding for Flatten in Flink Runner
- Open
- is cloned by
-
BEAM-12681 Support transcoding for Flatten in Samza Runner
- Open