[darcs-users] Re: [darcs-devel] Bug in patch reversal

[David Roundy]

(Coming out of hibernation for this...)

On Fri, Jan 20, 2006 at 12:35:11AM +0000, Ian Lynagh wrote:
> On Fri, Jan 20, 2006 at 12:52:42AM +0100, Tommy Pettersson wrote:
> > I have narrowed down where this bug happens.  It is in the
> > optimization of apply_list in PatchApply.lhs.
> > 
> > I have not wrapped my brain around the code
> 
> IIRC this code will be obsolete when darcs uses the new patch format
> internally, so IMO it would be better to work on that rather than fixing
> this.

That might be best, but this is definitely something that needs to be fixed
soon.  If noone is actively working on rewriting the hunk handling, I'd say
it'd be best to take Tommy's crude patch, slap the some
--set-scripts-executable code back in (I don't see any reason why the
simple approach would be incompatible with that feature), and apply it.
Then we would have a working darcs, so the tests that catch this bug could
be added to the test suite, and we wouldn't need to hurry on rewriting the
hunk handling.

And sorry to be so slow on the conflictor stuff.  Between job interviews
and Monica (my new girlfriend), I've been more than a little preoccupied.
I guess it's really Monica's fault...  I do have some ideas on conflictors
to work out, it's just that I don't seem to find time to do so.  I do plan
on getting back into serious darcs work, but at the moment don't seem to be
able to concentrate or think very clearly about much at all.

I guess this is a good time to thank Juliusz and Tommy again for taking
charge, and the rest of you for your contributions.  Your work helps keep
me from feeling too guilty about neglecting darcs, and keeps development
coming.  I'd especially like to thank Eric Kow--not because his
contributions are more valuable than others, but because he started
contributing since I went into seclusion, so I haven't had the opportunity
to thank him for a patch.
-- 
David Roundy
http://www.darcs.net