On Sun, 2004-02-08 at 22:03 +0100, guenther wrote:
> Gah. Sure, when redirecting mails, the other person acts like he does
> not exist and the mail gets directly to you. After all, there should not
> be a difference, if a human or a .forward file redirects mail... ;)

There should be a difference -- RFC2822 says Resent-* headers should be
added if a human does it, but not if a .forward file does it. See Â3.6.6

> Well, agreed. Any of this filters can fail in some weird situations. I
> should not have claimed no failures at all.

Actually, filtering on Sender: can indeed give the same false positives
as List-Id: if the MUA of the person resending the message actually
implements RFC2822 Â3.6.6 properly and keeps the same Sender: field when
resending a message.

What's better is to use the SMTP reverse-path -- which will often be
found in the Return-path: header. Although I use that myself for all
lists, I tend not to recommend it because different MTAs may not add it,
or may do something different. 

-- 
dwmw2


_______________________________________________
evolution maillist  -  [EMAIL PROTECTED]
http://lists.ximian.com/mailman/listinfo/evolution

Reply via email to