[gnu.emacs.gnus] Re: cc: nobody a bug?

[gnu.emacs.gnus] Re: cc: nobody a bug?

[Toby Speight]

[-- Attachment #1: Type: text/plain, Size: 53 bytes --]

Here's something I posted on Usenet.  Any thoughts?


[-- Attachment #2: Type: message/rfc822, Size: 1621 bytes --]

From: Toby Speight <Toby.Speight@streapadair.freeserve.co.uk>
Subject: Re: cc: nobody a bug?
Date: 15 Nov 1999 16:32:13 +0000
Message-ID: <u1z9riulu.fsf@lanber.cam.citrix.com>

Jochem> Jochem O. Huhmann <URL:mailto:joh@gmx.net>

0> In <URL:news:m3ln83md8g.fsf@nova.revier.com>, Jochem wrote:

Jochem> My humble opinion: Any string, be it "none" "never" or "nobody"
Jochem> *can* be a totally valid mail-adress on a local system and at
Jochem> least "nobody" in fact often *is*.  So an empty Mail-Copies-To
Jochem> seems to be the only working indicator for not Cc'ing at all.

AFAIK, "never" and "always" are the only allowed "magic" values.
But the concept of a "magic" value is flawed, as you say.  The fact
that this is news, and so local-system addresses are useless, can be
our rescue though.  I suggest that Gnus should treat any address
containing '@' in the usual way, otherwise it should check its list
of magic values.  If not found there, it should be brought to the
attention of the user (perhaps by putting some notice in the message
itself, which could be left in to warn the original poster of the
invalid value).

For example:

  [Your headers contained "Mail-Copies-To: nobody".]
  [I'm assuming you meant "Mail-Copies-To: never". ]

Re: [gnu.emacs.gnus] Re: cc: nobody a bug?

[Hrvoje Niksic]

Toby Speight <Toby.Speight@streapadair.freeserve.co.uk> writes:

> I suggest that Gnus should treat any address containing '@' in the
> usual way, otherwise it should check its list of magic values.

Since none of the magic values contains `@', that's exactly what Gnus
does.  What am I missing?

Re: cc: nobody a bug?

[Toby Speight]

Hrvoje> Hrvoje Niksic <URL:mailto:hniksic@iskon.hr>

0> In article <9t9zowdb3cz.fsf@mraz.iskon.hr>, Hrvoje wrote:

Hrvoje> Toby Speight <Toby.Speight@streapadair.freeserve.co.uk> writes:

>> I suggest that Gnus should treat any address containing '@' in the
>> usual way, otherwise it should check its list of magic values.

Hrvoje> Since none of the magic values contains `@', that's exactly
Hrvoje> what Gnus does.  What am I missing?

Has this changed since 0.95?

I see in `message-insert-to':

  (let ((co (message-fetch-reply-field "mail-copies-to")))
    (when (and (null force)
               co
               (or (equal (downcase co) "never")
                   (equal (downcase co) "nobody")))
      (error "The user has requested not to have copies sent via mail")))

and in `message-reply':

  (let (mct never-mct)
    (save-restriction
      (message-narrow-to-head)
      (setq mct (message-fetch-field "mail-copies-to"))

      ;; Handle special values of Mail-Copies-To.
      (when mct
        (cond ((or (equal (downcase mct) "never")
                   (equal (downcase mct) "nobody"))
               (setq never-mct t)
               (setq mct nil))
              ((or (equal (downcase mct) "always")
                   (equal (downcase mct) "poster"))
               (setq mct (or reply-to from)))))

      ;; ...
      (widen))
    ;; ...
    )


So Gnus appears to use only the four magic values "never", "nobody",
"always", and "poster".


Ah, I think I now understand your question.  What you are missing is
the bit where I say that if the value does not contain '@' and is not
found among the magic values, we show it to the user and ask whether
to mail a copy (adding a note in the message as we do so).  The reason
you missed this bit is, of course, because I forgot to write it -
thanks for pointing out the omission.

Re: cc: nobody a bug?

[Lars Magne Ingebrigtsen]

Toby Speight <Toby.Speight@streapadair.freeserve.co.uk> writes:

> Ah, I think I now understand your question.  What you are missing is
> the bit where I say that if the value does not contain '@' and is not
> found among the magic values, we show it to the user and ask whether
> to mail a copy (adding a note in the message as we do so).  

I don't think so.   People post on local news servers, where mail
addresses like "larsi" may be totally valid.

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: cc: nobody a bug?

[Toby Speight]

Lars> Lars Magne Ingebrigtsen <URL:mailto:larsi@gnus.org>

0> Toby Speight <Toby.Speight@streapadair.freeserve.co.uk> writes:

>> ... if the [Mail-Copies-To] value does not contain '@' and is not
>> found among the magic values, we show it to the user and ask whether
>> to mail a copy (adding a note in the message as we do so).

0> In <URL:news:m3ogctt1xh.fsf@quimbies.gnus.org>, Lars wrote:

Lars> I don't think so.  People post on local news servers, where mail
Lars> addresses like "larsi" may be totally valid.

Well, I for one would prefer to be asked (perhaps it would be nice if
Gnus first checked whether the address matches From or Reply-To; even
better if it looked to see if it's in the BBDB).  I understand that
not everyone would want that - something we can configure?

Re: cc: nobody a bug?

[Lars Magne Ingebrigtsen]

Toby Speight <Toby.Speight@streapadair.freeserve.co.uk> writes:

> Well, I for one would prefer to be asked (perhaps it would be nice if
> Gnus first checked whether the address matches From or Reply-To; even
> better if it looked to see if it's in the BBDB). 

I don't see the need for or utility of this.

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: cc: nobody a bug?

[Toby Speight]

Lars> Lars Magne Ingebrigtsen <URL:mailto:larsi@gnus.org>

0> In <URL:news:m3r9h6ljqa.fsf@quimbies.gnus.org>, Lars wrote:

Lars> Toby Speight <Toby.Speight@streapadair.freeserve.co.uk> writes:

>> Well, I for one would prefer to be asked (perhaps it would be nice if
>> Gnus first checked whether the address matches From or Reply-To; even
>> better if it looked to see if it's in the BBDB).

Lars> I don't see the need for or utility of this.

If the address doesn't contain "@", I don't think Gnus can determine
reliably whether it's

a) a local system address

b) a misspelling of "never"

c) a misspelling of "always"

So it takes a human to decide.  The only human my Gnus normally has
available is me.

Of course, other people will prefer Gnus to always assume either (b)
or (c).

Re: cc: nobody a bug?

[Lars Magne Ingebrigtsen]

Toby Speight <Toby.Speight@streapadair.freeserve.co.uk> writes:

> If the address doesn't contain "@", I don't think Gnus can determine
> reliably whether it's
> 
> a) a local system address
> 
> b) a misspelling of "never"
> 
> c) a misspelling of "always"
> 
> So it takes a human to decide.  

Yes.  And the human will see that it is Cc'ing "nevair" and remove
that (or not).

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen