Re: Small patch to enable use of gpg-agent with pgg

[Simon Josefsson <jas@extundo.com>]

Reinar, I think pgg is ready for v5-10 now.  I am sorry for all these
rapid changes.

Simon Josefsson <jas@extundo.com> writes:

>>> I think it is too complicated.  It is simpler to remove the function.
>>> Sascha, what do you think?  People who enable pgg-gpg-use-agent
>>> manually might be expected to make sure the agent really is running.
>>
>> I would like to see pgg do the right think when ever it can -- could
>> we keep it for emacs 22 and just not use it if its not available?  
>>
>> How about testing for the emacs version, or availability of
>> make-network-process?
>>
>> I'll purpose a more compatible patch later -- but I don't have the
>> time right now, sorry.
>
> Given that this function probably won't be used at all if Daiki's idea
> is implemented, I think we should remove the function for now.
> Daiki's approach (if I understand it correctly) of not second-guessing
> what parameters gpg will need is more reliable than this approach,
> even if this approach would work fine on all emacsen.

I removed the function.  The end result is this:

1) People who don't use the agent don't have to do anything, this is
   as before.

2) People who use the agent must enable pgg-gpg-use-agent manually.
   This is better than before, because the agent didn't work at all
   with pgg then.

3) People who use the agent must make sure the agent is still running
   (or disable the variable temporarily, or re-start the agent and
   fixing the environment variable using M-x setenv).

If/when Daiki's idea (as I understand it) is implemented, the end
result will be:

*) Pgg will invoke gpg and then figure out whether a passphrase is
   needed and only at that point query for it.  This will work
   regardless of whether the agent is used, whether a passphrase is
   required or not, or whether a pin code is required.  The user
   doesn't have to fiddle with any variable.

/Simon