Message-ID: <4DADF1F8.4030809@copiesofcopies.org>
Date: Tue, 19 Apr 2011 16:35:04 -0400
From: Aaron Williamson <aaron@copiesofcopies.org>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.9.2.15) Gecko/20110303 Lightning/1.0b2 Thunderbird/3.1.9
MIME-Version: 1.0
To: Sebastian Spaeth <Sebastian@SSpaeth.de>
Subject: Re: python documentation
References: <87hbb4zyp6.fsf@servo.finestructure.net>
	<87pqpsh0sg.fsf@SSpaeth.de>	<87mxjn68q7.fsf@eve.chaoflow.net>
	<877har7jcp.fsf@SSpaeth.de>	<87hb9v61qy.fsf@eve.chaoflow.net>
 <87oc4257t9.fsf@SSpaeth.de>
In-Reply-To: <87oc4257t9.fsf@SSpaeth.de>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Cc: Notmuch Mail <notmuch@notmuchmail.org>
Precedence: list

On 04/19/2011 04:01 PM, Sebastian Spaeth wrote:
> On Mon, 18 Apr 2011 17:03:01 +0200, Florian Friesdorf <flo@chaoflow.net> wrote:
>> Is it sane to have python bindings that need updates or would a plain
>> wrapper of the notmuch script (that maybe needs less updates) be better?
>> I don't want to imply that this can be answered with yes/no, but more to
>> get a comparison of the two approaches.
> 
> Invoking notmuch the script and piping the results? I wouldn't call that
> python bindings anymore then. Also it is bound to be much more expensive
> than directly interfacing libnotmuch.so.
> 
>> Naïvely, I imagine there is a header file and the generation of the
>> python bindings happens rather automagic.
> 
> I wish it would be that easy. :-)

I've never actually used it, so I may be way off base, but would it be possible
to use SWIG (http://www.swig.org/) to automatically generate bindings when the C
library is updated, or at least provide a decent starting point?

Aaron