[ 
https://issues.apache.org/jira/browse/CASSANDRA-18717?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17757551#comment-17757551
 ] 

Maxim Muzafarov commented on CASSANDRA-18717:
---------------------------------------------

[~samt] Thank you for sharing the link to the work, the last thing we want to 
do is to add another layer of complexity to the long-awaited work during the 
rebase. I have my own personal experience with keeping the distributed snapshot 
patch up to date for a long period of time during the review, and I remember 
how painful it can be.

Now I'm confused by the question of what criteria we should use to put one of 
the patches on hold and continue with the others. The number of changed files 
doesn't seem to be a good one. Should we introduce such criteria and publish 
them?

I have cherry-picked the fix of javadocs to the branch you provided, and 
actually, it's really easy to apply.  There were a few conflicts and ~8 errors 
with javadoc compilation. The cause of the conflicts was mostly because the 
method signature was changed e.g. TokenRingUtils#getPrimaryRangesForEndpoint, 
the last ones were caused by unacceptable symbols <, > or & e.g. the javadoc 
for the constant TCM_RECENTLY_SEALED_PERIOD_INDEX_SIZE. So, I think the 
CASSANDRA-18314 would probably be a bigger problem to rebase on and that's why 
I am confused.

The changes for your branch are here:
https://github.com/apache/cassandra/compare/cep-21-tcm...Mmuzaf:cassandra:cep-21-tcm-javadoc




> Fix errors or remove ant javadoc task
> -------------------------------------
>
>                 Key: CASSANDRA-18717
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-18717
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Documentation/Javadoc
>            Reporter: Ekaterina Dimitrova
>            Assignee: Maxim Muzafarov
>            Priority: Normal
>             Fix For: 5.0.x, 5.x
>
>          Time Spent: 50m
>  Remaining Estimate: 0h
>
> As discussed on CASSANDRA-17687, the javadoc task completes successfully, but 
> there are errors that deserve to be fixed if we keep the task around.
> If we do not plan to use it, we should remove it and reduce the clutter. 



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

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

Reply via email to