[ On Saturday, December 6, 2003 at 16:18:44 (-0800), Pat Lashley wrote: ]
> Subject: Re: [Exim] cyrus NUL failures
>
> > As for trying to handle VERP-based sender addresses, well good luck.
> > You'll fail far more often than you succeed! You can do no better than
> > play a game of catchup.
>
> What is your basis for that contention? Do you know of any mailing
> lists that use VERP but don't use the '*<localpart>=<domain>@...' form?
If I'm not mistaken this "format" isn't even standardized, and certainly
I've seen a large number of commercial mailing lists operating "in the
wild" that don't come even close to matching this form.
> In any case, it is no more of a game of catchup than any other aspect
> of attempting to recognize that the sender is a mailinglist address.
Well, there are the de facto standards outlined and documented for
Sendmail and all the other MTAs that work like Sendmail, then there are
the rest. If I'm not mistaken Sendmail still, unfortunately, tops the
list at over 70% of the working mailers on the internet, and certainly
the vast majority of mailers on the internet work like Sendmail does in
this respect (including Smail, and if I'm not mistaken Exim can handle
mailing lists just like Sendmail too if one bothers to try, as could
Postfix).
> I was under the impression that the 8BITMIME extension had relaxed
> that to allow message bodies with almost any octet to be passed
> unencoded. (Obvious exceptions including NUL, CR, and LF.) Assuming,
> of course, that both ends of the transfer support it.
8BITMIME is not a standard part of SMTP, not even in RFC 2822. RFC 3030
(defining 8BITMIME and BINARYMIME) is an optional extention only, and as
you say it's something that must be negotiated at connection time.
Standard SMTP messages _MUST_ be 7-bit only from beginning to end.
--
Greg A. Woods
+1 416 218-0098 VE3TCP RoboHack <woods@???>
Planix, Inc. <woods@???> Secrets of the Weird <woods@???>
