mengxr commented on pull request #31496:
URL: https://github.com/apache/spark/pull/31496#issuecomment-775599608


   @tgravescs I think there are two separate questions:
   
   1. Does it block 3.1 release? I agree with you that it shouldn't. It is 
indeed too late in the release process and the proposed changes are not 
critical bugs.
   2. Are the current APIs final? I think they are not. That is why they are 
labeled Evolving or DeveloperApi. So we still want to hear feedback from users, 
discuss them, and improve the APIs until we feel confident to mark them as 
stable.
   
   So how about the following (none blocking 3.1 release):
   
   1. In this PR, we complete the ScalaDoc and hide `ResourceAllocator` that 
should be marked private in 3.0. Seems everyone is okay with those changes.
   2. I hope we can agree on the following minor changes:
   
   * `Seq` -> `Array`: 
https://github.com/apache/spark/pull/31496/files#diff-319a3f0dfd7de6045eb11ad960180230c47e6f52b1b27d3c9a1f2d72f1615d9dR284
   * `build` -> `build()` 
https://github.com/apache/spark/pull/31496/files#diff-a6d96a65d9905b310451b125acac6610ffbd6b4548461bd1d5a18dc29282814aR71
   * Remove unused (and untested?) methods from builder: 
https://github.com/apache/spark/pull/31496/files#r571388509 and 
https://github.com/apache/spark/pull/31496/files#r571388779
   
   If we manage to get the above changes into 3.1 in time, I think it would 
improve API clarify/consistency. Then we can discuss other proposed 
renaming/refactoring after. Does it sound good?


----------------------------------------------------------------
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:
us...@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

Reply via email to