Hi Mark, > -----Original Message----- > From: Mark Ritchie [mailto:[EMAIL PROTECTED] > Sent: Wednesday, October 29, 2003 9:46 AM > To: [EMAIL PROTECTED] > Subject: RE: [SAtalk] Exessive HTML Code > > > > > Do you really think it would be a problem if we found more > than 3 instances of <i></i> in each email to mark it as spam? > Maybe I could just score it lower per instance, say .2 > > There were 58 instances of <i></i> in this email and 63 > instances of <b></b>.
I would go with 0.5 to a 1.0. You will need to break down every iteration into separate rules to make this work. There are no accumulation rules. So even if you had x number of instances of a match, you would only get one score not one score matching x times. So in the end you will wind up having the score of the rule, not score multiplied by x. You probably saw Jennifer Wheelers rules for testing this. I'll send you a copy if you did not see them. --Larry > -----Original Message----- > From: Larry Gilson > Sent: Wednesday, October 29, 2003 9:35 AM > To: Mark Ritchie; [EMAIL PROTECTED] > Subject: RE: [SAtalk] Exessive HTML Code > > > Yes, this would be possible. > > describe MY_RBDY_EXSV_TAG MY: Excessive HTML Tags > rawbody MY_RBDY_EXSV_TAG /<[bi]><\/[bi]>/i > score MY_RBDY_EXSV_TAG 4.0 > > Backhair did not hit because the number of characters within > the tag is fewer than 6. Creating rules to match fewer than > 6 characters within the tag delimiters creates false > positives. You will most certainly need to score it how you > want rather than the arbitrary number I supplied. > > --Larry ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
