Re: [PATCH v2 1/1] commit: display advice hints when commit fails

[Jonathan Tan <jonathantanmy@google.com>]

> > Junio, what are your plans over what you have in your tree? If you'd
> > like to hear Heba's opinion on it, then she can chime in; if you'd like
> > a review, then I think it's good to go in.
> 
> On hold until anything like those happens ;-) 
> 
> A random reviewer mentioning something on a patch (either in a
> line-by-line critique form or "how about doing it this way instead"
> counterproposal form) without getting followed up by others
> (including the original author) is a stall review thread, and it
> does not change the equation if the random reviewer happens to be me.

OK :-)

> > And I think the answer to that is "s" is used throughout the function in
> > various ways (in particular, used to print statuses both to stdout and
> > to the message template) so any wrapping or corralling of scope would
> > just make things more complicated. In particular, the way Heba did it in
> > v2 is more unclear - at the time of setting s->hints = 0, it's done
> 
> You mean "less clear" (just double checking if I got the negation right)?

Yes, less clear - v2 is less clear than v1.

> I think I've merged it to 'next' yesterday, but it does not mean
> that much as we are in -rc and it is not such an urgent "oops we
> broke it in this cycle, let's fix it" issue.  If we see a v3 that
> improves it, I do not mind at all reverting what I merged to 'next'
> and use the updated one instead (either way, it will be in 'master'
> during the next cycle at the earliest).

Sounds good - thanks.