[darcs-devel] bram cohen's new diff algorithm

[David Roundy]

On Tue, May 01, 2007 at 09:23:27PM +0200, Eric Y. Kow wrote:
> I noticed this post on planet revisioncontrol:
>   http://bramcohen.livejournal.com/37690.html
> 
> Bram Cohen has come up with what he says is a better diff algorithm, that
> it has made it into bazaar, and that he thinks all version control
> systems should switch.
> 
> Any comments?  Is this relevant to us?

We could certainly use it.  It gives somewhat bigger diffs and sometimes
runs slower than the diff algorithm we use, but supposedly gives more
human-friendly diffs, which might also help us to avoid unnecesary
conflicts.  Or might not.

It'd be (potentially?) interesting to see implemented.  I'd probably want
to keep our current code in place--not necesarily as an optional algorithm,
but because it'd probably be better for binary diffs, and (unless
modifications are made) would almost certainly be better for a
character-based diff, which is something I'd definitely like to have,
sometime in the next year or two.
-- 
David Roundy
Department 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-devel/attachments/20070502/fe264b9e/attachment.pgp