[darcs-users] darcs & git

[David Roundy]

On Mon, Nov 13, 2006 at 07:40:30PM +0100, Gour wrote:
> On Mon, 2006-11-13 at 03:21 +0100, Juliusz Chroboczek wrote:
> 
> > I do intend to fix Git support in Darcs.  It's a rather big job,
> > though -- the way we convert Git branches into Darcs branches is
> > broken by design, and David has explained to me how to do it right at
> > last FOSDEM.
> 
> Is it actually worth to do the job then?
> 
> It looks kernel won't anyway switch to darcs :-(

I think it is.  It's a theoretically valuable conversion, and will stress
test the functionality of the internal interfaces of darcs.  It may not be
worth the effort, depending on how much spare time Juliusz has, but it's
definitely a valuable work.  It also is nice in that it serves as a
prototype for true lossless interoperability with a differently-designed
RCS.  It may even be the case that the new conflict-handling idea will make
this relatively easy, as we'll need to introduce functions to work with
patches in a way that is closer to the ideal abstraction of sets.
-- 
David Roundy
Dept. of Physics
Oregon State University
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : http://lists.osuosl.org/pipermail/darcs-users/attachments/20061113/3445c310/attachment.pgp