[darcs-devel] Re: [darcs-conflicts] bit more of a summary of new conflict-handling ideas

[David Roundy]

On Wed, Jul 12, 2006 at 08:53:57PM +0100, Ganesh Sittampalam wrote:
> Some random comments about the bits I can digest easily:
> 
> On Wed, 12 Jul 2006, David Roundy wrote:
> 
> >2. We're going to have to be doing considerably more rearranging and
> >modifying of patches than in current darcs, in which a patch file is
> >untouched after it's created in a repository.
> 
> Is this true of optimize --reorder? If not, are gets already unsafe in 
> that particular case?

Indeed, gets and pulls are already unsafe if optimize --reorder is run
at the same time.  Odds may be good that you'd get a crash rather than
corruption, but there's no guarantee.

> >5. (The crazy idea) The new scheme is going to have to treat primitive
> >patches as the "first-class" objects, rather than named composite
> >patches as is currently the case.
> 
> This is a good thing anyway. Conflicts/dependencies with small parts of 
> composite patches causing the whole lot to cause problems (e.g. inability 
> to unpull a dependend-on patch) is a real nuisance sometimes.
>
> >I've now got a new related idea, which has a very strong appeal. How 
> >about we make the "name" (patch id, or PatchInfo) of a patch no longer 
> >be part of its identity, but instead be a sort of tag that's attached to 
> >it?  So that a given primitive patch could now have more than one name.
> 
> This sounds good.

Good.  I'm really getting attached to this
multiple-name-per-primitive-patch idea.  Of course, I'm sure you've
noticed that I get attached to ideas pretty quickly, and sometimes
end up abandoning them just as quickly.

> >This would give us "for free" the feature that patches that are 
> >identical except in name do not conflict.
> 
> I've said this before, but please don't allow this to happen without human 
> confirmation that it should. I know that in 99% of cases it's what's 
> wanted, but in the other 1% or 0.1% it's not, and it's much harder to sort 
> out the mess when it's happened unwantedly than vice versa.

With the implementation I've got in mind, we could easily add a
warning, perhaps even something analogous to the "!" in darcs changes
-s when this happens.  It shouldn't in any case be hard to query the
repository for primitive patches with multiple names.
-- 
David Roundy