[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Debian XSF SVN to git migration



On Thu, Jan 04, 2007 at 07:57:08AM +0100, Thierry Reding wrote:
> * David Nusinow wrote:
> > On Mon, Dec 25, 2006 at 02:16:23PM +0100, Michel Dänzer wrote:
> > > On Sat, 2006-12-23 at 15:29 -0500, David Nusinow wrote:
> [...]
> > After discussing this with Michel on irc, it doesn't look like it'll be
> > possible to use git the way I had written down. I've revised the policy
> > based on what Michel and I discussed[0]. 
> > 
> > It's much simplified, and basically behaves as you expect it to. If you
> > want to work on bleeding edge stuff from upstream, just pulling from the
> > debian* branch should give you the packaging. We believe it won't overwrite
> > local changes if you've cloned from, say, freedesktop.org because the
> > history should be intact.
> > 
> > The one weird thing is that there's no master branch by default, nor is
> > there an upstream branch, both of which git-buildpackage expects by default.
> > It's trivial to locally create those branches depending on what you're doing
> > though.
> > 
> > I'm going to wait a little while longer to see what everyone thinks. I want
> > to make sure everyone is back from holidays and had a chance to chew this
> > over before we make the move.
> >
> >  - David Nusinow
> > 
> > [0] http://wiki.debian.org/XStrikeForce/git-usage
> 
> The usage page still mentions that upstream changes should be cherry-picked
> into the master branch which we won't be using anymore. I'm guessing
> cherry-picks should go into upstream-* branches instead of the debian-* ones?

Either way is fine by me. The reason I had decided to go with the debian
branch was to keep the upstream as close to a linear history as possible. I
had envisioned the upstream branch being more like a released version. I
guess that having cherry picks go to the upstream* branch is more intuitive
though, since everyone seems confused about it, so let's go with that. I'll
update the wiki tonight.

 - David Nusinow



Reply to: