The way I've set mail routing in our company is that I wrote a Ruby script that the mail gets piped into before going through dspam. All DNS zones and Email addresses are LDAP based. Which gives me a bunch of configuration option. I can turn off or on spam filtering for a certain domain (my actual dspam setup is done on per-domain basis, instead of per-user basis) via a our admin control center. And our clients can whitelist/black list individual senders or sender domains by using their client control center utility.
When a sender is whitelisted, dspam is skipped and mail is just delivered. I cannot copy/paste the script but if you are interested in doing it this way then I could give you a few insightful hints. Petr
