Re: disable "org-decrypt: auto-save-mode may cause leakage"

[Pieter Praet <pieter@praet.org>]

[-- Attachment #1: Type: text/plain, Size: 2301 bytes --]

On Sun, 28 Aug 2011 09:02:58 +0200, Achim Gratz <Stromeko@nexgo.de> wrote:
> Pieter Praet <pieter@praet.org> writes:
> > Apparently not config-related.
> 
> So it does happen with just "emacs -Q"?  I cannot reproduce this.

Well, not quite... :) Some minimal config is required.

1. Run EMACS like this:
   (don't forget to adjust the load-path)

   #+begin_src sh
     emacs -Q --eval="(progn
         (setq debug-on-error t)
         (add-to-list 'load-path \"~/src/org-mode/lisp\")
         (require 'org-install)
         (require 'org-crypt)
         (org-crypt-use-before-save-magic))"
   #+end_src

2. Open the ECM (in att) and complete the TODO's in sequence.


When running `org-clock-goto', the user will be prompted:
: org-decrypt: auto-save-mode may cause leakage. Disable it for current buffer? (y or n)
... which is rather uncalled for in this context.


I we'd be calling an error @ lisp/org-crypt.el::170, we'd get the following
backtrace upon running `org-clock-goto':

#+begin_example
  Debugger entered--Lisp error: (error "blah")
    signal(error ("blah"))
    error("blah")
    (progn (error "blah") (cond (... ... ...) (... ...) (... ... ...) (t nil)))
    (if buffer-auto-save-file-name (progn (error "blah") (cond ... ... ... ...)))
    (when buffer-auto-save-file-name (error "blah") (cond (... ... ...) (... ...) (... ... ...) (t nil)))
    org-decrypt-entry()
    run-hooks(org-reveal-start-hook)
    org-reveal()
    (let* ((recent nil) (m ...)) (org-pop-to-buffer-same-window (marker-buffer m)) (if (or ... ...) (widen)) (goto-char m) (org-show-entry) (org-back-to-heading t) (org-cycle-hide-drawers (quote children)) (recenter) (org-reveal) (if recent (message "No running clock, this is the most recently clocked task")) (run-hooks (quote org-clock-goto-hook)))
    org-clock-goto(nil)
    call-interactively(org-clock-goto nil nil)
#+end_example


So, either org-crypt shouldn't be imposing itself on `org-reveal-start-hook'
(see lisp/org-crypt.el::250) XOR `org-clock-goto' shouldn't call `org-reveal'.
I'll let others decide...


Peace


> Regards,
> Achim.
> -- 
> +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+
> 
> Wavetables for the Waldorf Blofeld:
> http://Synth.Stromeko.net/Downloads.html#BlofeldUserWavetables
> 
> 

Peace

-- 
Pieter


[-- Attachment #2: crypt-vs-clock.org --]
[-- Type: text/plain, Size: 466 bytes --]

* NOTE here's an encrypted entry (symmetric, pass = "test")              :crypt:
-----BEGIN PGP MESSAGE-----
Version: GnuPG v1.4.10 (GNU/Linux)

jA0ECgMCBqqZSbotsitg0nwBKep+l0L85gNUFsipCPc6Nn/VKiRoZrbBU37f0+7w
yTBvfSXMpBKc+D8TZ4ZoIPwqNwc52TZA4phh5VQIs5T96PAXYH6yAQIBbG7ggMWV
4zrU6NiHU/dCCznsbObQXGkMg1tVU1Kp6A9EapGBklh24mcrwhG3BZQcqzN2
=uLcl
-----END PGP MESSAGE-----

* TODO run `org-clock-in' (`C-c C-x C-i') here

* TODO run `org-clock-goto' (`C-c C-x C-j') here