as threatened, here is a complete patch against the current gnu.org
lisp/gnus repository.  i've migrated all the allout encryption
functionality to pgg and exercised this stuff pretty substantially,
and am very comfortable with the way it works.

there are a bunch of fixes, as well as extensions so the passphrases
can be managed by the calling functions, as well as refinement of the
passphrase caching mechanism (so multiple timeouts don't pileup for a
single passphrase - the last one replaces the prior ones).   lotsa
goodies.  the patch includes sascha wilde's symmetric-encryption
additions - basically, the whole deal.

what it does not include is revision of the pgg-pgp.el and pgg-pgp5.el
modules, to track changes to the pgg api.  i started in on that, only
to discover that they were already broken by the addition of a 'sign'
parameter to some of the encryption functions.  there's no way to
repair that short of implementing the feature for the other packages,
and i'm not in a position to do that.  (plus, i'm not sure i am
comfortable with the way it was done for pgg-gpg.)

so it looks like pgg-pgp.el and pgg-pgp5.el are disfunctional already,
without my changes.  i'm hoping that's not going to prevent the
introduction of my changes.

i'm going to be heading offline for five days, starting tomorrow, so
it'll be a bit before i can respond to questions.  i hope the patch
i'm sending (which  includes detaild ChangeLog entries) will be
helpful to anyone that wishes to navigate them.

and once again, i'm not sure the attachment will be delivered this
time.  (it wasn't on a recent try, but then was on a subsequent
message.  this is the first awkward gmail bug i've hit...)

ken
ken.manheimer@gmail.com

On 10/10/05, Ken Manheimer <ken.manheimer@gmail.com> wrote:
> for those of you following the developments at home (:-), here's an
> incremental patch on top of what i sent out a few days ago.  i fixed a
> small stack of bugs in pgg-gpg.el that settles my complaint about
> prompting with the secret key identity, and also filled in a small
> oversight in the changes i sent out a few days ago.
>
> while this could be the last pgg patch, i'm still developing, hence
> the incremental.  i plan to send out a full patch when i've finished
> transitioning allout to pgg, at which point i expect my mucking with
> pgg to be settled.  so, repository maintainers may want to wait for
> that, while i'm hoping those actively involved (eg, sascha) will
> scrutinize and, ideally, exercise these patches.
>
> we'll see if the attached patch makes it through this time.  (i
> haven't re-attached it, which i think was the problem last time.)  if
> not, look forward to a followup...