Re: Searchable Mail Archives

skip(a)pobox.com writes: Thanks for the offer! They're already mbox. However, this really isn't a bottleneck. I have the scripts to do it using procmail, and for the period in question this is acceptably accurate. The delay here is more with (a) space, which is somewhat scarce on the list host, and (b) breaking handcoded links into the archives both in past posts and on the website. What I'd rather have, if you would, is advice on integrating SpamBayes into the Mailman pipeline. Recently some of the newer spams are getting past both SpamAssassin (SpA in the diagram below) and my own procmail filters, and updating them is a very painstaking and unreliable process. I'd like to replace them with something more automated. Currently our process looks like this: --> MTA --> procmail --> Mailman --> | ^ v | SpA kind of a Rube Goldberg arrangement, but it has the advantage that most spam never gets to Mailman. Would it make sense to call SpamBayes from procmail? How does one train SpamBayes? _______________________________________________ XEmacs-Beta mailing list XEmacs-Beta(a)xemacs.org http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta