[darcs-users] Coalescing patches

Nik darcs at babel.homelinux.net
Wed Sep 23 14:28:55 UTC 2009

Hi Stephen,

Thanks for your input.

Stephen J. Turnbull wrote:
> Nik writes:
>  > http://nik.homelinux.net/files/darcs-coalescing.html
>  > 
>  > If anyone has any comments on this, I would be very interested to read them.
> At the commit level, I don't really see how you can beat git, since
> for a pure "squash", it can edit the history DAG directly, and in
> general can accomplish rebase very efficiently.

I was focusing more on the user interface than the underlying storage.
I had inferred that git and darcs would be as effective as the other in 
the actual coalesce operation.
True, git can efficiently edit the history DAG, but with Darcs there is 
(effectively) no history DAG to edit, so the point is moot, isn't it?
> What would be interesting would be if patch theory allowed Darcs to
> coalesce better (ie, with fewer conflicts, or spurious changes dragged
> along with the desired ones),

I assumed that the coalesce was combining a number of smaller patches 
into a single larger one. Darcs has the patch contents and the patch 
dependency information which is sufficient to ensure the coalesced patch 
is correct. Correct?

>  and/or the new hunk editing capabilities
> allowed coalescing to have an efficient UI at a lower lever than
> commits.

So you're talking more of the situation in which some of the patches to 
coalesce contain multiple updates, some of which should be coalesced and 
some not?
Is this also the cause of the "conflicts or spurious changes" you talk 
of in the point above?

So, assuming that the storage model and record semantics are unchanged, 
you don't see any real benefit to darcs users in the suggested change?


More information about the darcs-users mailing list