From: Iustin Pop <iusty@k1024.org>
Date: Sun, 09 Mar 2008 10:31:35 GMT
Subject: Re: [PATCH 2/2] Fix git pull handling of the quiet option
Message-ID: <20080309103135.GF32019@teal.hq.k1024.org>
In-Reply-To: <7v8x0ssfud.fsf@gitster.siamese.dyndns.org>

On Sat, Mar 08, 2008 at 08:04:26PM -0800, Junio C Hamano wrote:
> Iustin Pop <iusty@k1024.org> writes:
> 
> > Although git pull has a documented quiet option,...
> 
> I think that is a documentation bug.  pull accepts all options for fetch
> for the sole purpose of passing them intact to underlying fetch, and some
> options to fetch does not even make much sense in the context of pull.
> 
> Also options to pull needs to come first; the options pull does not know
> about is a signal for pull that the rest is for consumption of underlying
> fetch.
Ah, I see. This is indeed not clear from the documentation.

> If you want to teach --quiet to pull, however, your patch is the right
> approach.  pull would eat --quiet and make a note for itself, and passes
> that to underlying fetch (and perhaps merge).
> 
> You also need to sign-off your patch and add tests to make sure that other
> people will not break your enhancement in the future.
Thanks, so the approach would be:
  - resend git merge patch including tests
  - and then resend git pull patch, again including tests.

Thanks for the feedback, I will try to see how the unittests are written
and hopefully come back with some more patches.

iustin