[darcs-users] Fwd: Towards a conflict-free revision control system.

[Ganesh Sittampalam]

On Wed, 21 Jan 2009, Declan Naughton wrote:

>> ... and then there is the fun that David Roundy liked to point out: how
>> many of us program in complete compilable (and thus
>> deconstructable/reconstructable) code *all the time*?  What if you want
>> to save a non-working fragment in progress?  What if you make a tiny
>> mistake and the file fails to parse correctly?
>
>
> But, in this case couldn't darcs reverting to its current behavior?
>
> The state of the repository could be like:
>
> patch A we have this function that function blah
> patch B more of the same
> patch C oh no we can't parse it like that no more but well we have
> this DIFF now.

The problem is that the diff and the more semantic-based patches probably 
couldn't be merged at all. So you'd end up always in conflict with other 
trees that didn't have the parse problem.

Ganesh