Re: PGG and 'pgg-sign' command.

[Simon Josefsson <jas@extundo.com>]

Jorge Godoy <godoy@ieee.org> writes:

> Simon Josefsson <jas@extundo.com> writes:
>
>> Jorge Godoy <godoy@ieee.org> writes:
>>
>>> I'm trying to use 'pgg-sign', but although it makes the screen blink
>>> and repositions the cursor at the beginning of the screen, nothing
>>> happens and the message isn't signed. When I use pgg-sign-region with
>>> a marked region, everything works just fine.
>>
>> The patch below should fix it, but it is ugly and the same solution
>> doesn't work for pgg-encrypt.  Elisp Q: How do I interactively call a
>> function specifying the parameters directly?
>>
>> --- pgg.el.~6.9.~	2002-10-30 19:40:41.000000000 +0100
>> +++ pgg.el	2002-11-10 09:06:30.000000000 +0100
>> @@ -228,7 +228,10 @@
>>  If optional arguments START and END are specified, only sign data
>>  within the region."
>>    (interactive "")
>> -  (pgg-sign-region (or start (point-min)) (or end (point-max)) cleartext))
>> +  (save-excursion
>> +    (set-mark (or start (point-min)))
>> +    (goto-char (or end (point-max)))
>> +    (call-interactively 'pgg-sign-region)))
>>    
>>  ;;;###autoload
>>  (defun pgg-verify-region (start end &optional signature fetch)
>
> I guess that the solution pointed out by Josh Hubber
> (<87znsil4pb.fsf@mail.paradoxical.net>) solves the problem and even
> make this command useless...

pgg-encrypt/pgg-decrypt/pgg-sign/pgg-verify should now display output
when called interactively.  I'm not sure they are useless, users that
aren't familiar with Emacs often dislike *-region functions.  And it
saves power users a key press to set the mark too.

> I know that you added them because of some prior message of mine,
> but what do you think about mapping 'pgg-sign' and 'pgg-encrypt' to
> those mml- equivalent options?

I don't think that is a good idea.  PGG should IMHO only be a
low-level interface to OpenPGP implementations such as GnuPG.  If you
need more high-level operations, you can use a more high-level
interface like MML.