Re: [PATCH/RFC] git-svn: don't create master if another head exists

[Marcin Owsiany <marcin@owsiany.pl>]

On Tue, Jun 26, 2012 at 03:03:07PM -0700, Junio C Hamano wrote:
> Marcin Owsiany <marcin@owsiany.pl> writes:
> 
> > diff --git a/git-svn.perl b/git-svn.perl
> > index 0b074c4..2379a71 100755
> > --- a/git-svn.perl
> > +++ b/git-svn.perl
> > @@ -1612,9 +1612,9 @@ sub post_fetch_checkout {
> >  		}
> >  	}
> >  
> > -	my $valid_head = verify_ref('HEAD^0');
> > +	return if verify_ref('HEAD^0');
> >  	command_noisy(qw(update-ref refs/heads/master), $gs->refname);
> 
> Given that your original motivation was "I do not want master, I am
> using something else for my primary branch", I change that still
> shows "update-ref refs/heads/master" smells like sweeping something
> under the rug

I'm not so sure... With this change, git-svn will only create master on
the initial "clone" and I think that's fine. It's consistent with what
"git clone" does when cloning a regular git repository.

It seems that I have slightly misinterpreted git-svn's actions in my
initial post in 2009. I thought it always updated "master" to the
most recent upstream commit. In reality it only every _creates_ it at
the most recent commit. But never fast-forwards it if it pre-exists.

This makes my idea to do the same to "my something else instead of
master" much less attractive. In fact I don't think such behaviour would
be useful.

I think with the suggested patch git-svn works as I would like it to:
 - creates "master" at initial checkout - consistent with git clone
 - using a different "tracking-like" branch is possible with "dcommit"
 - does not re-create "master" on fetch - so the annoying part is gone

-- 
Marcin Owsiany <marcin@owsiany.pl>              http://marcin.owsiany.pl/
GnuPG: 2048R/02F946FC  35E9 1344 9F77 5F43 13DD  6423 DBF4 80C6 02F9 46FC

"Every program in development at MIT expands until it can read mail."
                                                              -- Unknown