[darcs-users] Identical patches and dependencies

[David Roundy]

On Wed, Aug 13, 2008 at 2:05 PM, Ganesh Sittampalam <ganesh at earth.li> wrote:
> On Wed, 13 Aug 2008, David Roundy wrote:
>>> Ah, I see this is talked about on
>>> http://wiki.darcs.net/index.html/DarcsTwo
>>>
>>> I must have read it some time ago and forgotten about it.
>>>
>>> Anyway, from where I'm sitting, this just shows that primitive patches
>>> must be named [...],
>>
>> No, it doesn't show any of that.  It just shows that we need to fix
>> darcs, because it's buggy.
>
> To me, the suggestion on that page of "atomization" and giving predictable
> names to the resulting patches is strong evidence that naming primitive
> patches is the right thing to do.

It's absolutely the wrong thing for me to do, I can't afford just now
to start discussing a reworking of the patch theory for darcs 3.0.
Nor can I afford to fix the bugs in darcs 2, nor do I know of anyone
with the inclination and knowledge to do so.  Which is why I have
strongly recommended that people consider not using darcs.

In any case, discussion of what should have been done or could have
been done is useless and a waste of my time.  Discussion of how to fix
bugs in darcs without a complete rewrite is also a waste of my time,
but at least it's not useless.

David