[GitHub] [incubator-sedona] jiayuasu edited a comment on pull request #557: [SEDONA-67] Support Spark 3.2

[GitBox <gi...@apache.org>]

jiayuasu edited a comment on pull request #557:
URL: https://github.com/apache/incubator-sedona/pull/557#issuecomment-955576126


   The failed test cases are caused by the Spark 3.2.0 change in UnaryExecNode and BinaryExecNode. So technically, Sedona join extends a different ExecNode in Spark 3.2.0.
   
   Spark < 3.2.0 UnaryExecNode and BinaryExecNode implementation:
   
   ![image](https://user-images.githubusercontent.com/10948864/139554566-039438b3-d5ce-49ad-84ab-1184032ca1cf.png)
   
   ![image](https://user-images.githubusercontent.com/10948864/139554739-b4c5ac56-d6ba-4c2e-81cb-0032de47c6e0.png)
   
   Spark 3.2.0 UnaryExecNode and BinaryExecNode implementation
   
   ![image](https://user-images.githubusercontent.com/10948864/139554810-d5904226-f817-4c71-98e1-394048f13691.png)
   
   ![image](https://user-images.githubusercontent.com/10948864/139554824-adca11d0-6f11-4f5b-9247-d20ff580c8ce.png)
   
   This is related to https://github.com/apache/incubator-sedona/pull/558
   
   However, on Sedona's side, this does not require any actual code change.
   
   @Kimahriman @Imbruced Can we figure out a solution for this? In the worst case, we have to cut a separate release that is compiled on Spark 3.2+. In short, we have to release Sedona for Spark 2.4, 3.0 (3.0-3.1), 3.2. This is a headache...
   


-- 
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.

To unsubscribe, e-mail: dev-unsubscribe@sedona.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org