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

[Sebastien Vauban]

Hi Pieter,

Pieter Praet wrote:
> On Tue, 19 Jul 2011 01:12:23 +0200, Bastien <address@hidden> wrote:
>> "Sebastien Vauban" <address@hidden> writes:
>> 
>> > I simply call C-c C-x C-j to jump to the current clocking entry, and the
>> > question "org-decrypt: auto-save-mode may cause leakage. Disable it for
>> > current buffer? (y or n)" pops up -- while the current buffer has nothing
>> > special, as said above.
>>
>> Weird.
>> 
>> Maybe you have some hook in `org-clock-goto-hook' that makes Org visit some
>> buffer containing encrypted entries?
>
> Apparently not config-related.
>
> I don't use org-clock.el (and as such don't have any org-clock related
> config, `org-clock-goto-hook' is nil), but due to a recent freak accident
> involving spastic typing, I've managed to trigger the issue. Apologies for
> dismissing it earlier, Sebastien.

No problem, Pieter. I did not expand the thread further by lack of time, and
-- at this point in time -- I can't prove yet whether you're wrong or right
(saying it could be config-related) in my case.

> `org-clock-goto' erroneously causes `org-decrypt-entry' to be called, though
> it's not immediately obvious how or why this would happen.

Your analysis and patch seems right to me, just from reading it.

However, in my case, I have the question coming up even when
org-clock-goto'ing to/from a buffer that does not contain any :crypt: entry!
But, as stated, by lack of time, I simply can't try to reduce the problem to
an ECM right now. So, I'll first test your patch, when pushed to the repo, and
report any new observation.

Best regards,
  Seb

-- 
Sebastien Vauban