On Wed, Aug 03, 2005 at 06:12:52PM -0700, Professional Software
Engineering wrote:
At 01:34 2005-08-04 +0200, Dallman Ross wrote:
On panix, X-Envelope-To: always contains the envelope-To. The
MTA is not sendmail.
Then obviously it isn't restricted to the same logic which
sendmail happens to use. I don't happen to agree with the way
sendmail does it - in the header insertion phase (same phase
where the local Received: is tacked on), if there are multiple
envelope recipients, sendmail doesn't emit that data because
it'd be a security issue - revealing other recipient addresses.
Sendmail isn't set up to add headers LATER, say at local delivery
time, by which point it has already discerned the local userid.
I see. Thanks for explaining its dumbness. :-)
I'm curious - if you have multiple aliases at your ISP which
point to the same mailbox, and two are addressed on a message,
do you get TWO copies of the message (each with a separate
X-Envelope-To:), or ONE?
I see two. I believe that is how it should be. But in order for
that to happen, I had to ask the administrator to turn of a
brain-dead duplicate-Message-ID "swallower" "feature" that was
eating my redirected mail without obvious explanation. Thankfully,
the admins complied.
If X-Envelope-To: is empty when the message was sent to multiple
local recipients, what's the point of even having it? I mean,
one can simply look for the lowest "for user@" string in the
Received chain,
... which coincidentally won't exist if there were multiple local
recipients.
Yes, I thought that was my point. :-) I think you knew that.
I'm stating so for others.
Dallman
____________________________________________________________
procmail mailing list Procmail homepage: http://www.procmail.org/
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail