Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
From: Justus Winter <4winter@informatik.uni-hamburg.de>
User-Agent: alot/0.3+
To: Mark Walters <markwalters1009@gmail.com>,  notmuch@notmuchmail.org
References:
 <1332291311-28954-1-git-send-email-4winter@informatik.uni-hamburg.de>
	<1332291311-28954-2-git-send-email-4winter@informatik.uni-hamburg.de>
	<87fwcopu5g.fsf@qmul.ac.uk>
In-Reply-To: <87fwcopu5g.fsf@qmul.ac.uk>
Message-ID: <20120416215112.15615.94985@thinkbox.jade-hamburg.de>
Subject: Re: [PATCH 1/7] Split notmuch_database_close into two functions
Date: Mon, 16 Apr 2012 23:51:12 +0200
Precedence: list

Quoting Mark Walters (2012-03-31 19:17:15)
> Secondly, I think the patch series could be made clearer and easier to
> review. If you do it in three steps
> =

> 1) change of notmuch_database_close to notmuch_database_destroy (just
>    the function name change)
> 2) split the new notmuch_database_destroy into two as in the current
>    first patch
> 3) Make any changes (if there are any) of notmuch_database_destroy to
>    notmuch_database_close.
> =

> The advantage is that the first change is easy to test (essentially does
> it build) and then changes from notmuch_database_destroy to
> notmuch_database_close in step 3 are explicit rather than the current
> situation where we need to grep the code to see if some instances of
> notmuch_database_close were not changed to notmuch_database_destroy.

I don't buy it. The patch series first touches the library and
documentation and the lib compiles fine. The next patch updates the
cli tools, all of them compile fine afterwards.

Every patch addresses the issue component wise, this seems rather
natural for me.

Cheers,
Justus