Tony L. Svanstrom wrote: > On Fri, 24 May 2002 the voices made Matt Sergeant write: > > >>Tony L. Svanstrom wrote: >> >>> Would it be possible to set up an environment where you can test changed/new >>>rules against a set number of spam and non-spam, preferably the same ones used >>>for the current GA-scores? >> >>The best way is to just extract the header from emails you know break, >>and write tests for them. It's dead easy, just: >> >>test RULE_NAME ok Something this rule should match >>test RULE_NAME fail Something this rule should not match >> >>Unfortunately it doesn't cope with embedded newlines in headers yet, but >>I'm hoping that will become completely irrelevant once SA does proper >>email parsing. > > > I was thinking that it'd be a great help to get the GA-generated score as a > sign of if one's heading in the right direction; esp. helpful when you think > you might be able to improve an existing rule.
Probably the best way to do this is to have something that can check a body of spam and non-spam to see how many emails a particular rule matches. The GA weights mainly on this value. It'd probably be pretty easy to copy t/rule_tests.t for this, which sets all the scores to zero except the one you're interested in. Matt. _______________________________________________________________ Don't miss the 2002 Sprint PCS Application Developer's Conference August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
