Hi, SPARK-18591 <https://issues.apache.org/jira/browse/SPARK-18591> might be a solution to your problem but making assuming in your UDAF logic on how Spark will process the aggregation is really a risky thing. Is there a way to do it using Windows function with ORDER BY clause to enforce the processing in this case?
----- Nattavut Sutyanyong | @nsyca Spark Technology Center http://www.spark.tc/ -- View this message in context: http://apache-spark-developers-list.1001551.n3.nabble.com/Aggregating-over-sorted-data-tp19999p20206.html Sent from the Apache Spark Developers List mailing list archive at Nabble.com. --------------------------------------------------------------------- To unsubscribe e-mail: dev-unsubscr...@spark.apache.org
