Gnus development mailing list
 help / color / mirror / Atom feed
From: Simon Josefsson <jas@extundo.com>
Cc: ding@hpc.uh.edu
Subject: Re: I have to enter my passphrase twice?
Date: Mon, 05 May 2003 19:22:47 +0200	[thread overview]
Message-ID: <ilur87dpau0.fsf@latte.josefsson.org> (raw)
In-Reply-To: <871xzdnz1s.fsf@pooh.honeypot.net> (Kirk Strauser's message of "Mon, 05 May 2003 11:22:39 -0500")

Kirk Strauser <kirk@strauser.com> writes:

> At 2003-05-05T15:55:29Z, Simon Josefsson <jas@extundo.com> writes:
>
>> Do you use GCC?  Perhaps the message that is mailed is signed correctly,
>> but the second signature (for GCC) fails, and the GCC code doesn't signal
>> an error.  Can you look in your GCC group?
>
> Yes, I use GCC.  Out of curiosity, why would it sign the message twice?

Because the body in GCC may be encoded differently than the body that
is sent via mail.  If this sounds weird, here is an elaboration:

Consider if you set gnus-gcc-externalize-attachments and attach a file
to your message.  Then the body encoded for GCC is very different than
the body encoded for mail.  (Namely, the latter one includes the file,
whereas the former only includes a external MIME reference.)

If you posted your message to a newsgroup too (that is, both via mail
and news, and GCC), you might have to sign the message three times,
since Usenet have different encoding requirements than mail (e.g.,
iso-8859-1 may be prefered over utf-8 in one newsgroup, and the other
way around in another, and any choice may be different from what is
prefered for mail).

In general, a given MML message encoded for mail, news or GCC do not
necessarily look the same.  So they need different signatures.

I guess the security system could "cache" (unsigned-msg, signed-msg)
tuples so the user is not queried twice if the encodings are
identical.  But this sounds like work, and might end up not being used
in the majority of cases anyway due to subtle differences.

Perhaps you object to these ideas, and want GCC to simply save a
_copy_ of what was actually mailed.  Currently this is not what GCC
does, but if you want that behavior instead, BCC yourself and filter
them into your sent-mail folder.

> One lingering question, though: where did it get "kirk@strauser.com" as my
> key ID?  That happens to be a userid on the key I use, but I never specified
> it anywhere.

It was taken from the From: header.




  reply	other threads:[~2003-05-05 17:22 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-05-05 14:30 Kirk Strauser
2003-05-05 14:56 ` Simon Josefsson
2003-05-05 15:02 ` Kirk Strauser
2003-05-05 15:55   ` Simon Josefsson
2003-05-05 16:22     ` Kirk Strauser
2003-05-05 17:22       ` Simon Josefsson [this message]
2003-05-05 19:11         ` Kirk Strauser
2003-05-05 19:32           ` Simon Josefsson
2003-05-05 16:38 ` Steve Youngs
2003-05-05 21:25   ` Kirk Strauser

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=ilur87dpau0.fsf@latte.josefsson.org \
    --to=jas@extundo.com \
    --cc=ding@hpc.uh.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).