From: Sven Verdoolaege <skimo@liacs.nl>
Date: Sat, 11 Aug 2007 06:45:55 GMT
Subject: Re: [PATCH resend] git-apply: apply submodule changes
Message-ID: <20070811064555.GC29996@liacs.nl>
In-Reply-To: <7vd4xupqwh.fsf@assigned-by-dhcp.cox.net>

On Fri, Aug 10, 2007 at 10:43:42PM -0700, Junio C Hamano wrote:
> Sven Verdoolaege <skimo@kotnet.org> writes:
> >  	else if (patch->old_name) {
> >  		size = xsize_t(st->st_size);
> >  		alloc = size + 8192;
> >  		buf = xmalloc(alloc);
> > -		if (read_old_data(st, patch->old_name, &buf, &alloc, &size))
> > +		if (S_ISGITLINK(patch->old_mode))
> > +			size = snprintf(buf, alloc,
> > +				"Subproject commit %s\n", sha1_to_hex(ce->sha1));
> > +		else if (read_old_data(st, patch->old_name, &buf, &alloc, &size))
> 
> Who guarantees that ce is given to apply_data() in this codepath?

Oops.  I guess it shows that I only tested it through git-rebase.
Would changing

		if (check_index) {

on line 2093 to

		if (check_index || S_ISGITLINK(patch->old_mode)) {

be acceptable?  Adding a conditional call to read_cache(), of course.
We're not going to be able to apply submodule patches without
an index, anyway.
Or should we just refuse to apply submodule patches if --index
has not been specified?

skimo