In comp.mail.sendmail 73blazer <73blazer@yoyoma.com>:
> I've been searching around, and I can't find anything relavent.
> I'm new to sendmail language, and run a small family webserver/mailserver.

> I wish to implement and per-user blacklist, that is if UserA says he
> doesn't want to recieve mail from address spam@spam.com or domain
> spam.com, I don't want his blacklist to block mails going to UserB.

Generally not a sendmail issue. There are several possibilities,
mailscanner as already pointed out might be just what you want,
it can integrate SA (SpamAssassin) and AV scanner(s) into
receiving/delivering mail with various MTA.

http://www.mailscanner.info/

Mailwatch is a http frontend and should allow to set up per user
blacklist, if you have those user local on the MTA:

http://mailwatch.sourceforge.net/

Iirc you can teach per user bayes with it, which is a safer
approach then just blacklisting.

In general tossing spam as early as possible (rejecting during
smtp time) is the best way. One of the best approaches today
needing minimal resources would be greylisting. If possible
adaptive during smtp time, though I am not aware of any other MTA
then exim (+sa-exim) providing adaptive greylisting.

Fighting spam needs careful MTA + SA setup, to keep legitimate
mail flow with lowest user impact possible. False positive are
hardly avoidable for a large enough user base these days. YMMV

Strange enough 80% of all spam comes from a few well known
people, there seem just zero political volition to stop it. You
can find those dudes, most with pictures if you check out the
ROSKO list:

http://www.spamhaus.org/

Good luck

--
Michael Heiming (X-PGP-Sig > GPG-Key ID: EDD27B94)
mail: echo zvpunry@urvzvat.qr | perl -pe 'y/a-z/n-za-m/'
#bofh excuse 456: Noobs anywhere...