I thought about this issue as well and originally considered the same perspective that you took below before discarding it in favor of the ongoing position argument that MD is a framework that should be capable of any milter task that swayed me.
I use a similar argument with SpamAssassin. SpamAssassin's strength is that it is an excellent scoring framework for anti-spam. If I create an algorithm that rocks for anti-spam (or pro-ham), I can simply integrate it into the existing framework as SA with an appropriately weighted score. This framework and reusability is what makes programs like MD and SA invaluable to me. I will state though that processing the helo prior to filter_sender would cause astronomical False Positive rates for my server. I even know people using helo processing in filter_end because in their setup, they would have FPs under their setup. So I would not suggest a recommended milter implement filter_helo but otherwise, it's a perfectly valid idea under the auspices of increasing the flesh on the framework capabilities. My $0.02, KAM > The regular filter sender is hadling the 'helo' quite well and it only > a command > away from the 'helo' command...?!? so what is the cost benefit? > > 10x David and other for a great tool! _______________________________________________ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list [email protected] http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
