mar...@mejor.pl wrote:
Hi! Thanks to AXB seek-in-phrases-in-log works OK. Now I'm on the next step with automated creating rules. I suspect that mk_meta_rule_scores doesn't assign scores correctly. I set in mk_meta_rule_scores: my %scoremap = ( '70' => '1.5', '4' => '2.0', '0.01' => '3.0', );
If I understand correctly (quite possible I don't; I haven't dug in to the internals of this stage), the %scoremap hash above indicates the percentage of the messages that need to hit on a subrule for that subrule to be included in one of the meta rules.
So, 70% of the mail in the set would need to hit on a subrule for it to be included in the first group, 4% in the second, and 0.01% in the third.
If I read the information flow correctly, this is actually decided by seek-phrases-in-log, which spits out subrules that reached a certain hit rate in blocks, followed by the "# passed hit-rate threshold nnn" line. mk_meta_rule_scores just takes that in, collects the rule names in each block, and spits out the meta.
(In fact the stock setup refers to mk_meta_rule, although it's nearly identical to mk_meta_rule_scores.)
By raising the hit percentage to 70%, you're requiring that 70% of the spam you're using must hit on one of the subrules. TBH, by that point, you may as well hand-extract a couple of the subrules and make them static, standalone rules.
My experience has been that you need lots of mail to generate multiple metas in any case; I've taken a different tack and separated out different groups of mail for different generated rule sets instead.
-kgd
