Hi Jonathan,
Jonathan Nieder writes:
> Almost there. ÂTo comfort overly-worried people like me that think we
> have not finished converted all die() calls yet, wouldn't this need to
> look like
>
> Â Â Â Âpersist_head(head);
> Â Â Â Âwhile ((commit = ...)) {
> Â Â Â Â Â Â Â Âint status_or_error;
>
> Â Â Â Â Â Â Â Â/*
> Â Â Â Â Â Â Â Â * Checkpoint. ÂIf do_pick_commit exits, make sure the user
> Â Â Â Â Â Â Â Â * can still use "git cherry-pick --continue" to recover.
> Â Â Â Â Â Â Â Â */
> Â Â Â Â Â Â Â Âpersist_todo(revs.commits, opts);
>
> Â Â Â Â Â Â Â Âstatus_or_error = do_pick_commit(...);
> Â Â Â Â Â Â Â Âif (status_or_error)
> Â Â Â Â Â Â Â Â Â Â Â Âreturn status_or_error;
> Â Â Â Â}
>
> Â Â Â Â/* Success! */
> Â Â Â Âremove_todo(opts);
> Â Â Â Âremove_head();
> Â Â Â Âreturn 0;
You told me about this on IRC last night, but I forgot to squash the
relevant patch into this series. Thanks :)
> And with that, this would no longer depend on the (valuable enough on
> their own terms) patches 1 and 2 so they could be treated as a
> separate series, no?
Yes, but I'd like to keep 1 and 2 in this series, if that's alright.
Another series will have to take care of error handling more
extensively.
With a few tests, would it be useful to get this series merged in? It
implements '--abort', which is already useful.
-- Ram
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html