hadoop-pig-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Olga Natkovich" <ol...@yahoo-inc.com>
Subject RE: guideline on branching
Date Tue, 04 Dec 2007 02:06:34 GMT
Yes, our plan is to go with option a), thanks!


-----Original Message-----
From: Torsten Curdt [mailto:tcurdt@apache.org] 
Sent: Monday, December 03, 2007 4:38 PM
To: pig-dev@incubator.apache.org
Subject: Re: guideline on branching

On 03.12.2007, at 23:35, Olga Natkovich wrote:

> The reason we want to make a branch is because the changes to be made 
> for the project are going to take the whole system apart. While a 
> person or a group of people working on this project, they can commit 
> it to the branch as they see fit. Once they are ready to commit the 
> changes, then we can treat this branch as a big patch and take it 
> through the same process as we do with patches. So the people working 
> on the branch may or may not be committers.

There are a couple of ways dealing with this.

a) Have a committer create a feature branch and then have the
individuals provide patches that a committer will review and apply.  
If there are enough substantial patches there is enough evidence to
invite him to become a committer. He then can happily continue on the
branch ...or whereever.

b) Have that person/group use git/git-svn and then provide the "big
patch" you are talking about.

I would be in favor of a) as that will properly show better the
individual contributions. But bottom line is: committing is for
committers only.


View raw message