From: Austin Clements <amdragon@MIT.EDU>
To: Mark Walters <markwalters1009@gmail.com>, notmuch@notmuchmail.org,
	Adam Wolfe Gordon <awg+notmuch@xvx.ca>
Subject: Re: [PATCH v3 0/5] emacs: show: lazy handling of hidden parts
In-Reply-To: <1370024806-6616-1-git-send-email-markwalters1009@gmail.com>
References: <1370024806-6616-1-git-send-email-markwalters1009@gmail.com>
User-Agent: Notmuch/0.15.2+83~g8bee3c4 (http://notmuchmail.org) Emacs/23.4.1
	(i486-pc-linux-gnu)
Date: Fri, 31 May 2013 19:33:15 -0400
Message-ID: <87wqqe2090.fsf@awakening.csail.mit.edu>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Precedence: list

LGTM other than the tiny things I mentioned.  As Mark says, there are a
lot of paths through this code (though he did a good job of minimizing
them), so it'll need some testing.  Of course, what better place to test
than git master?

As a heads up, I think this will have a trivial conflict with
id:"1369876428-13537-4-git-send-email-amdragon@mit.edu", which is
currently in the ready queue.

On Fri, 31 May 2013, Mark Walters <markwalters1009@gmail.com> wrote:
> This is a substantially reworked version of
> id:1369555061-21361-1-git-send-email-markwalters1009@gmail.com
> attempting to answer the (very constructive) criticism of Austin.
>
> The changes are: move the insert-part-header call from the
> insert-part-text/plain function to insert-bodypart in line with all
> the other part insertion functions. 
>
> The other change in patch 2 is that we no longer need to pass
> declared-type down to the part insertion functions so we remove this
> argument.
>
> The actual invisibility code has been substantially reworked. There
> were two main aims: to remove the insertion of dummy text and to make
> sure that toggling a lazy-part which cannot be rendered automatically
> calls the default action for such a part (e.g. saving or viewing).
>
> I have split this code into 3 pieces: two pieces of simple code
> movement as preparation and then the new code to do the actual
> lazy part handling.
>
> All tests pass and everything I can think of to try works (viewing
> pdf, lazy rendered html, lazy rendered pdf, toggling all types,
> repeated toggling) but there are a lot of cases so testing is
> especially helpful.
>
> Best wishes
>
> Mark
>
>
>
>
>
> Mark Walters (5):
>   emacs: show: fake wash parts are handled at insert-bodypart level
>   emacs: show: move the insertion of the header button to the top level
>   emacs: show: pass button to create-overlays
>   emacs: show: modify the way hidden state is recorded.
>   emacs: show: implement lazy hidden part handling
>
>  emacs/notmuch-show.el |  183 +++++++++++++++++++++++++++++--------------------
>  1 file changed, 107 insertions(+), 76 deletions(-)
>
> -- 
> 1.7.10.4