From: Dmitry Kurochkin <dmitry.kurochkin@gmail.com>
To: David Edmondson <dme@dme.org>, notmuch <notmuch@notmuchmail.org>
Subject: Re: [PATCH v2] emacs: Make the part content available to the
	mm-inline* checks.
In-Reply-To: <cunboq06szv.fsf@hotblack-desiato.hh.sledj.net>
References: <1326907993-11054-1-git-send-email-dme@dme.org>
	<1326908371-11949-1-git-send-email-dme@dme.org>
	<877h0o99aj.fsf@gmail.com>
	<cunlip47tir.fsf@hotblack-desiato.hh.sledj.net>
	<874nvs96ps.fsf@gmail.com>
	<cunboq06szv.fsf@hotblack-desiato.hh.sledj.net>
User-Agent: Notmuch/0.11+73~g8a5402e (http://notmuchmail.org) Emacs/23.3.1
	(x86_64-pc-linux-gnu)
Date: Thu, 19 Jan 2012 13:23:37 +0400
Message-ID: <87k44o6o6e.fsf@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Precedence: list

On Thu, 19 Jan 2012 07:39:32 +0000, David Edmondson <dme@dme.org> wrote:
> On Wed, 18 Jan 2012 23:00:15 +0400, Dmitry Kurochkin <dmitry.kurochkin@gmail.com> wrote:
> > > That would allow a sane default (("image/*" "text/*") perhaps), but also
> > > allow more to be added to that list (or some to be removed), either by
> > > code that detected the (in)ability to render it or the user.
> > 
> > Perhaps there is such a list in mm already?
> 
> There's a function which does almost exactly this - `mm-inlinable-p'. It
> has a list of types and tests, `mm-inline-media-tests'. Some of those
> tests require access to the part content to decide if the part is
> inlinable. Many of them don't. The image/jpeg test _does_ want access to
> the part content.
> 
> We're already using this function, of course.
> 
> `mm-inlined-p' is the corresponding "does the user want this part
> inlined" test. It's much simpler and never looks at the part content.
> 
> Currently we merge those tests into one:
> 
> 	(if (and (mm-inlinable-p handle)
> 		 (mm-inlined-p handle))
> 
> and have acquired part content either before or after the combined test.
> 
> Perhaps we could test `mm-inlined-p' first, then insert the content,
> then test `mm-inlinable-p'? That way we would not acquire the content
> for parts for which the user (or code) has selected not to inline the
> content.
> 

Makes sense to me.

> Currently `mm-inlined-p' suggests that the following should be inlined
> by default:
> 
>     "image/.*" "text/.*" "message/delivery-status" "message/rfc822"
>     "message/partial" "message/external-body" "application/emacs-lisp"
>     "application/x-emacs-lisp" "application/pgp-signature"
>     "application/x-pkcs7-signature" "application/pkcs7-signature"
>     "application/x-pkcs7-mime" "application/pkcs7-mime"
> 
> These are the only types for which we'd acquire the part content for
> examination by `mm-inlinable-p' (by default).

Sounds good.

Regards,
  Dmitry