From: Mark Walters <markwalters1009@gmail.com>
To: Austin Clements <amdragon@MIT.EDU>, notmuch@notmuchmail.org
Subject: Re: [PATCH 08/11] emacs: Support caching in
	notmuch-get-bodypart-{binary, text}
In-Reply-To: <1398105468-14317-9-git-send-email-amdragon@mit.edu>
References: <1398105468-14317-1-git-send-email-amdragon@mit.edu>
	<1398105468-14317-9-git-send-email-amdragon@mit.edu>
User-Agent: Notmuch/0.15.2+615~g78e3a93 (http://notmuchmail.org) Emacs/23.4.1
	(x86_64-pc-linux-gnu)
Date: Thu, 24 Apr 2014 11:46:28 +0100
Message-ID: <87sip3t37f.fsf@qmul.ac.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Precedence: list


On Mon, 21 Apr 2014, Austin Clements <amdragon@MIT.EDU> wrote:
> (The actual code change here is small, but requires re-indenting
> existing code.)
> ---
>  emacs/notmuch-lib.el | 52 ++++++++++++++++++++++++++++++----------------------
>  1 file changed, 30 insertions(+), 22 deletions(-)
>
> diff --git a/emacs/notmuch-lib.el b/emacs/notmuch-lib.el
> index fc67b14..fee8512 100644
> --- a/emacs/notmuch-lib.el
> +++ b/emacs/notmuch-lib.el
> @@ -503,33 +503,39 @@ (defun notmuch-parts-filter-by-type (parts type)
>     (lambda (part) (notmuch-match-content-type (plist-get part :content-type) type))
>     parts))
>  
> -(defun notmuch-get-bodypart-binary (msg part process-crypto)
> +(defun notmuch-get-bodypart-binary (msg part process-crypto &optional cache)
>    "Return the unprocessed content of PART in MSG as a unibyte string.
>  
>  This returns the \"raw\" content of the given part after content
>  transfer decoding, but with no further processing (see the
>  discussion of --format=raw in man notmuch-show).  In particular,
>  this does no charset conversion."
> -  (let ((args `("show" "--format=raw"
> -		,(format "--part=%d" (plist-get part :id))
> -		,@(when process-crypto '("--decrypt"))
> -		,(notmuch-id-to-query (plist-get msg :id)))))
> -    (with-temp-buffer
> -      ;; Emacs internally uses a UTF-8-like multibyte string
> -      ;; representation by default (regardless of the coding system,
> -      ;; which only affects how it goes from outside data to this
> -      ;; internal representation).  This *almost* never matters.
> -      ;; Annoyingly, it does matter if we use this data in an image
> -      ;; descriptor, since Emacs will use its internal data buffer
> -      ;; directly and this multibyte representation corrupts binary
> -      ;; image formats.  Since the caller is asking for binary data, a
> -      ;; unibyte string is a more appropriate representation anyway.
> -      (set-buffer-multibyte nil)
> -      (let ((coding-system-for-read 'no-conversion))
> -	(apply #'call-process notmuch-command nil '(t nil) nil args)
> -	(buffer-string)))))
> -
> -(defun notmuch-get-bodypart-text (msg part process-crypto)
> +  (let ((data (plist-get part :binary-content)))
> +    (when (not data)
> +      (let ((args `("show" "--format=raw"
> +		    ,(format "--part=%d" (plist-get part :id))
> +		    ,@(when process-crypto '("--decrypt"))
> +		    ,(notmuch-id-to-query (plist-get msg :id)))))
> +	(with-temp-buffer
> +	  ;; Emacs internally uses a UTF-8-like multibyte string
> +	  ;; representation by default (regardless of the coding
> +	  ;; system, which only affects how it goes from outside data
> +	  ;; to this internal representation).  This *almost* never
> +	  ;; matters.  Annoyingly, it does matter if we use this data
> +	  ;; in an image descriptor, since Emacs will use its internal
> +	  ;; data buffer directly and this multibyte representation
> +	  ;; corrupts binary image formats.  Since the caller is
> +	  ;; asking for binary data, a unibyte string is a more
> +	  ;; appropriate representation anyway.
> +	  (set-buffer-multibyte nil)
> +	  (let ((coding-system-for-read 'no-conversion))
> +	    (apply #'call-process notmuch-command nil '(t nil) nil args)
> +	    (setq data (buffer-string)))))
> +      (when cache
> +	(plist-put part :binary-content data)))
> +    data))

I am a little puzzled by this but that could be lack of familiarity with
elisp. As far as I can see plist-put will sometimes modify the original
plist and sometimes return a new plist. If the latter happens then I
think it works out as if we hadn't cached anything as the part passed to
the function is unmodified. That might not matter in this case (though I
find the lack of determinism disturbing).

Or is something else going on?

Best wishes

Mark



> +
> +(defun notmuch-get-bodypart-text (msg part process-crypto &optional cache)
>    "Return the text content of PART in MSG.
>  
>  This returns the content of the given part as a multibyte Lisp
> @@ -546,7 +552,9 @@ (defun notmuch-get-bodypart-text (msg part process-crypto)
>  	     (npart (apply #'notmuch-call-notmuch-sexp args)))
>  	(setq content (plist-get npart :content))
>  	(when (not content)
> -	  (error "Internal error: No :content from %S" args))))
> +	  (error "Internal error: No :content from %S" args)))
> +      (when cache
> +	(plist-put part :content content)))
>      content))
>  
>  ;; Workaround: The call to `mm-display-part' below triggers a bug in
> -- 
> 1.9.1
>
> _______________________________________________
> notmuch mailing list
> notmuch@notmuchmail.org
> http://notmuchmail.org/mailman/listinfo/notmuch