Without  --signoff, you already get set as the "committer", while the
author is maintained. You can verify this by running "coho last-week" and
see that it separates commits you wrote vs commits that you did from pull
requests. That said, adding "--signoff" couldn't hurt.

Squishing & fixing up does maintain authorship, so I think that's fine. I
also sometime clean up whitespace & tabs->spaces.

It's definitely nice to squash & fix the commit messages not just for
release notes, but so you can figure out what the commit does from the "git
log", and so that they can be reverted easily.

I've been doing this for several months now and haven't had anyone
complain, so I don't think those submitting the PRs care too much.



On Fri, Feb 28, 2014 at 5:51 PM, Michal Mocny <[email protected]> wrote:

> Does the squash keep original author info?  I know the hashes change so
> they don't match up, but if we have the author and a reference to the PR in
> the commit, I think thats fine for me.
>
> Alternative is to ask the contributor to do the squash, which we do try to
> do, but its usually the non-responsive contributors that submit the nasty
> PR in the first place (go figure).
>
> -Michal
>
>
> On Fri, Feb 28, 2014 at 5:14 PM, Joe Bowser <[email protected]> wrote:
>
> > Hey
> >
> > I saw the wiki was updated, and I'm not quite sure how I feel about this:
> > https://wiki.apache.org/cordova/ProcessingPullRequests
> >
> > # REPO_NAME example: "js"
> > # PULL_REQUEST_NUMBER example: "44"
> > curl
> >
> https://github.com/apache/cordova-REPO_NAME/pull/PULL_REQUEST_NUMBER.patch
> > | git am
> > git rebase origin/master -i
> >
> > First, I'd add git am --signoff so that it adds your git e-mail to it.
> >  This just looks nice, but I'm not sure how I feel about doing the
> > commit squishing and pruning, since we want to have a nice audit trail
> > back to GitHub to see what we were doing.  I know it makes it harder
> > to do release notes, but the one thing that I get super picky about is
> > who wrote what code, and it matching up.
> >
> > Am I just being all Apache about this, and we could be way more lax?
> > What are people's thoughts?
> >
> > Joe
> >
>

Reply via email to