Eric Kow <[EMAIL PROTECTED]> writes:
> This means, for example, that if you do a darcs pull and there are no
> patches to pull, the posthook does not get run.
>
> It makes sense to me, personally. Any objections, darcs-users?
As something of a by-stander, allow me to play devil's advocate:
- Can a post-hook currently determine if the command that invoked it was
a noop? If so, the post-hook can begin with "if noop, stop," and
opt-out of being run.
- With this change, post-hooks aren't run on noops. There's no way for
the post-hook to change this behaviour (opt-in), precisely because it
isn't run.
- Are there cases where a post-hook *should* run after a noop?
Regarding my third point, I can think of one case: an apply post-hook
that logs that a push attempt was made. That is,
apply post-hook logger -t darcs "Push attempt! Haxxor on ttyS0!"
_______________________________________________
darcs-users mailing list
[email protected]
http://lists.osuosl.org/mailman/listinfo/darcs-users
