On 4 Mar 2012, at 12:28, Ben Franksen wrote: > As far as I am concerned, the 'darcs send computes minimal context' solution > would be perfect. But I have no idea how difficult that is (both for the > darcs developers and then for darcs). For a compromise, maybe some > approximation of 'minimal' would be good enough in practice.
Ah yes, I think I remember that one issue is that computing the minimal context is actually expensive. I wonder what Darcs could compute that would a useful “smaller context”. Maybe since Darcs already bases the bundle off the latest shared tag between sender and trunk (don't quote me on this, just a guess), it could just be minimal-assuming-that-latest-tag? -- Eric Kow <http://erickow.com>
signature.asc
Description: Message signed with OpenPGP using GPGMail
_______________________________________________ darcs-users mailing list darcs-users@darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users