> An annoying problem of Mail.app is that there appears to be > no way to insert a header into a message already in Mail.app. > So I can't do things like untrain_from_header().
It's pretty rare to be able to modify a message once it's already in the mail client, whatever mail client you're using (without hacking into the stored file(s)). > It's infeasible to keep my own database to record what message has > been trained and what their verdicts are, since sooner or later I'll > run into scalability problem. Just clean out the database occasionally (besides which, this takes almost no space, and space is cheap). sb_filter will move to using the messageinfo database that the rest of SpamBayes uses at some point, FWIW. > So for now I just assume*every* incoming email is brand new, so I > won't even do an untrain() on them. As to the case where users > explicitly ask to re-classify some emails (via Mail.app's contextual > menu item "Apply Rules"), I'll just do a filter() without train(). > > And so far I haven't figured out an elegant way to let users > train SB in a batch mode. Any Mac dev out there - welcome to chime in. :-) What do you mean by "batch mode"? What exactly are you wanting users to be able to do? =Tony.Meyer _______________________________________________ spambayes-dev mailing list [email protected] http://mail.python.org/mailman/listinfo/spambayes-dev
