Details
-
Sub-task
-
Status: Resolved
-
Major
-
Resolution: Fixed
-
4.0.0
Description
Currently Spark Connect supports only UDFs. We need to add support for UDAFs, specifically `Aggregator[INT, BUF, OUT]`.
The user-facing API should not change, which includes Aggregator methods and the `spark.udf.register("agg", udaf(agg))` API.
Attachments
Issue Links
- links to