From: Pieter Praet <pieter@praet.org>
To: David Edmondson <dme@dme.org>, Austin Clements <amdragon@MIT.EDU>
Subject: Re: [PATCH] [RFC] emacs: merge overhauled
	`notmuch-cycle-notmuch-buffers' into `notmuch'
In-Reply-To: <cun1uqz50p4.fsf@hotblack-desiato.hh.sledj.net>
References: <cunpqek597g.fsf@hotblack-desiato.hh.sledj.net>
	<1326710551-23122-1-git-send-email-pieter@praet.org>
	<cunvcob52ku.fsf@hotblack-desiato.hh.sledj.net>
	<87wr8rc2ex.fsf@praet.org>
	<cun1uqz50p4.fsf@hotblack-desiato.hh.sledj.net>
User-Agent: Notmuch/0.11+78~g6c58370 (http://notmuchmail.org) Emacs/23.3.1
	(x86_64-unknown-linux-gnu)
Date: Mon, 16 Jan 2012 17:45:25 +0100
Message-ID: <87r4yza95m.fsf@praet.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: Notmuch Mail <notmuch@notmuchmail.org>
Precedence: list

On Mon, 16 Jan 2012 11:47:03 +0000, David Edmondson <dme@dme.org> wrote:
> On Mon, 16 Jan 2012 12:28:06 +0100, Pieter Praet <pieter@praet.org> wrote:
> > Perhaps we could use `ido-completion-help', like Org-mode does with
> > `org-iswitchb' ?  OTOH, that would require an extra [RET].  Bah...
> 
> I'm a no-ido person.
> 
> > > Hmm, yes, that seems reasonable.
> > > 
> > > > FYI: causes 2 compile warnings due to requiring the cl package at runtime.
> > > 
> > > `cl' is still `eval-when-compile' though - won't that break?
> > 
> > Haven't noticed any issues here, apart from the complaints @ compilation.
> 
> If you load the compiled lisp into 'emacs -Q' and then attempt to call
> the function, it doesn't complain? (I haven't tested.)

Good call !  No joy:
  Debugger entered--Lisp error: (void-function remove-if-not)

Should've known.

So let's just get rid of that cl requirement, shall we :)

Patch follows.


Peace

-- 
Pieter