hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Steve Loughran (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HADOOP-11731) Rework the changelog and releasenotes
Date Fri, 03 Apr 2015 12:29:54 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-11731?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14394362#comment-14394362

Steve Loughran commented on HADOOP-11731:

I don't think CHANGES.TXT works that well. We may think it does, but that's because without
the tooling to validate it, stuff doesn't get added and so it can omit a lot of work. Then
there's the problem of merging across branches, and dealing with race conditions/commit conflict
between other people's work and yours.

automated generation not only gives us the change list, it adds the option of generating a
hyperlinked version where you can actually click through the JIRAs.

I do recognise the concerns about JIRA naming and tagging —but that's a process thing that
is correctable. If we can put the effort in to maintaining CHANGES.TXT, we can do it for JIRAs.
Many use cases will be easier. For example commit something to trunk and then backport it,
today, is:
# pull trunk
# apply patch to trunk
# edit changes.txt
# commit
# attempt to push
# if unlucky: fix CHANGES.TXT conflict & commit, push.
# close JIRA

The backport workflow becomes
# switch to branch-2
# pull
# apply  the code bits of the patch
# edit changes.txt
# commit
# go to trunk
# fix changes.txt
# push
# remember to edit JIRA

Forward porting is easier: edit & commit branch-2, cherry pick the patch, push both

Now think of a JIRA-only workflow
# apply patch to trunk
# commit
# close in JIRA with 3.0.0 version

you'll only get commit conflict if someone patched the source files, so commits are more likely
to go through.

# cherry pick patch
# commit & push
# in JIRA, change fix version.

It's backporting & cross version code where this stuff really excels. 

> Rework the changelog and releasenotes
> -------------------------------------
>                 Key: HADOOP-11731
>                 URL: https://issues.apache.org/jira/browse/HADOOP-11731
>             Project: Hadoop Common
>          Issue Type: New Feature
>          Components: documentation
>    Affects Versions: 3.0.0
>            Reporter: Allen Wittenauer
>            Assignee: Allen Wittenauer
>             Fix For: 3.0.0
>         Attachments: HADOOP-11731-00.patch, HADOOP-11731-01.patch, HADOOP-11731-03.patch,
HADOOP-11731-04.patch, HADOOP-11731-05.patch, HADOOP-11731-06.patch, HADOOP-11731-07.patch
> The current way we generate these build artifacts is awful.  Plus they are ugly and,
in the case of release notes, very hard to pick out what is important.

This message was sent by Atlassian JIRA

View raw message