[darcs-devel] [patch1911] WIP: use Prim patches in rebase toedit

[Ganesh Sittampalam]

Ganesh Sittampalam <ganesh at earth.li> added the comment:

>> But patch1915 shows that it does actually cause a
>> regression, without a solution for suspending conflicted
>> patches into fixups directly.
> 
> You haven't mentioned it expressly but I guess this failure is fully
> expected. 

Right.

>   RebaseContainer prim ~ FL (Suspended prim)
>   Suspended prim ~ FL RebaseName :> Named (RebasePatch prim)
>   RebasePatch prim ~ FL (Fixup prim) :> prim
>   Fixup prim ~ prim
> 
> (where ~ means equivalence, not necessarily equality)
> 
> fullUnwind (we really should find a better name) should return
> 
>   RebasePatch prim :> FL (Fixup prim)

IMO fullUnwind would be independent of rebase and would continue to return

  FL prim :> prim :> FL prim

We can then build on it in the rebase code to get the types you mention.

Apart from that I roughly agree, except perhaps the names, but we can
review that later.

> I wouldn't care too much about compatibility of the on-disk
> representation, at least not in the initial effort. Just make sure that
> we keep enough of the old code that we can upgrade an existing rebase patch.

What I intend to do is produce a draft sequence of patches that gets us
to the rough end-state so we can discuss that, and then go back and
figure out how to keep backwards compatibility. I'm working on that at
the moment.

__________________________________
Darcs bug tracker <bugs at darcs.net>
<http://bugs.darcs.net/patch1911>
__________________________________