You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@hadoop.apache.org by Allen Wittenauer <aw...@linkedin.com> on 2010/03/11 19:46:20 UTC

Commit Log Hacks ( https://issues.apache.org/jira/browse/HADOOP-6628 )

Seeing as how our two non-mainline distributions don't have real release
notes published, I've built two quick and dirty perl scripts (yes perl! Fear
the ops people!) to take the commit log from CDH and the Yahoo! changes
files to spit out Confluence-style wiki output.  Changing it to other wiki
output shouldn't be too difficult.

99% of the time, it gets it correct, but sometimes weirdness happens and it
will take some manual cleaning to get it perfect.

Anyway, I thought these might be handy for other folks to help compare and
contrast the differences between the two.  Now it is pretty easy to see how
divergent a given distribution is from mainline, how hard it will be for
them to rebase onto mainline, etc.  Another interesting data point is
comparing the official bug description and the description in the commit
log....

Have fun.

[Files are attached to https://issues.apache.org/jira/browse/HADOOP-6628.  I
might improve them as we go along, but who knows for sure.  ]


Re: Commit Log Hacks ( https://issues.apache.org/jira/browse/HADOOP-6628 )

Posted by Eli Collins <el...@cloudera.com>.
Hey Allen,

CDH now includes Apache style release notes generated from the change
log. Hope this helps.

http://archive.cloudera.com/cdh/testing/hadoop-0.20.1+169.56.releasenotes.html

Thanks,
Eli

On Thu, Mar 11, 2010 at 10:46 AM, Allen Wittenauer
<aw...@linkedin.com> wrote:
>
> Seeing as how our two non-mainline distributions don't have real release
> notes published, I've built two quick and dirty perl scripts (yes perl! Fear
> the ops people!) to take the commit log from CDH and the Yahoo! changes
> files to spit out Confluence-style wiki output.  Changing it to other wiki
> output shouldn't be too difficult.
>
> 99% of the time, it gets it correct, but sometimes weirdness happens and it
> will take some manual cleaning to get it perfect.
>
> Anyway, I thought these might be handy for other folks to help compare and
> contrast the differences between the two.  Now it is pretty easy to see how
> divergent a given distribution is from mainline, how hard it will be for
> them to rebase onto mainline, etc.  Another interesting data point is
> comparing the official bug description and the description in the commit
> log....
>
> Have fun.
>
> [Files are attached to https://issues.apache.org/jira/browse/HADOOP-6628.  I
> might improve them as we go along, but who knows for sure.  ]
>
>