cloud-fan commented on PR #55878:
URL: https://github.com/apache/spark/pull/55878#issuecomment-4453318597

   @dongjoon-hyun Your suggestion makes me think what should be the rule of 
deciding the released versions. In the past, when we merge a PR to master only, 
the released version is the next feature release version (e.g. `4.1.0`), which 
implies that all releases after `4.1.0` will contains the commit. If the PR is 
also backported to a release branch, like `branch-4.0`, we additionally include 
the next maintenance release version (e.g. `4.0.2`), which indicates that 
releases before `4.0.2` do not have the commit.
   
   So the principle is to use a minimal set of released versions to tell users 
which Spark versions contain the commit. That being said:
   
   | Merge branches | released versions |
   |---|---|
   | `master` | `[5.0.0]` |
   | `master`, `branch-4.x` | `[4.3.0]` |
   | `master`, `branch-4.x`, `branch-4.2` | `[4.2.0]` |
   | `master`, `branch-4.x`, `branch-4.2` if 4.2.0 has released | `[4.3.0, 
4.2.1]` |
   | `master`, `branch-4.2` | `[5.0.0, 4.2.0]` |
   | `master`, `branch-4.2` if 4.2.0 has released | `[5.0.0, 4.2.1]` |
   
   The last 2 rows are mostly for completeness, but it should not happen in 
practise.
   
   How do you think of this principle?


-- 
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