JoshRosen commented on PR #47334:
URL: https://github.com/apache/spark/pull/47334#issuecomment-2226161408

   > After this PR, there is a chance of losing a binary compatibility at 
Apache Spark 3.5.2+ against Apache Spark 3.4.x, isn't there?
   
   I don't think so:
   
   - Assume that Spark 3.5.0 was binary compatible with Spark 3.4.x (except in 
the known cases we excluded).
   - Let's say we make a change in, say, 3.5.2. which breaks compatibility with 
Spark 3.4.x: because Spark 3.5.0 was compatible with 3.4.x, the 3.4.x methods 
or types against which we could be incompatible should all be present in a 
compatible fashion in Spark 3.5.0, so the types of 3.4.x incompatibilities that 
we want to prevent will also be 3.5.x incompatibilities. 
   
   I think that this argument is sound, but let me know if there is a corner 
case that I'm overlooking.


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to