On 7/3/06, Zachary P. Landau <[EMAIL PROTECTED]> wrote:
> >This feature has been requested often, but support for it doesn't
> >exist
> >yet. See http://otherbugs.darcs.net//index.html?q=37 for some
> >discussion about it. I did write a patch to implement it once but
> >there
> >are many issues with it. A real implementation will have to be
> >done in
> >a different way. I have some ideas for it but never got a chance
> >to try
> >it out.
>
> Outstanding for over 18 months? Yikes. I didn't see anything on
> the
> bug report regarding the issues that make this difficult.
I don't think it is that difficult. But I'm a lousy Haskell programmer
and don't have much free time. And as far as I know, nobody else has
attempted it.
That being said, it can be a little trickier than it first appears. The
main issue with my implementation was that it didn't account for
dependencies. If you ignore a patch, but one you want depends on it,
something sane should happen. It could be as simple as saying "you
can't pull this patch until you unignore this patch:" or it could have
an interactive mechanism for unignoring the patch. My implementation
will pull in the patch, not realizing it depends on another one that has
been ignored, which is not so good.
I think the implementation choices here include:
1. Automatically and silently ignore patches that depend on ignored patches
2. Present the user with the option of unignoring the old patch,
adding the new one to the ignore list, or canceling the operation
I like #2, because it gives the user notification of new patches he
may be missing but does not annoy him about them any more than
necessary to accomplish that.
_______________________________________________
darcs-users mailing list
[email protected]
http://www.abridgegame.org/mailman/listinfo/darcs-users
