> yeah, it means we can't write rules that look for the lack of '/' > characters, but in this case I'm more concerned about more closely > following what the browsers do with those URIs than the ability to > make a rule to catch the issue. if it becomes something that occurs > frequently by spammers, we can reinvestigate.
-1 We should return both the fixed and unfixed versions ... now. (Remember, this is a utility function not a rule to catch spam. It has a primary function of correctness and a secondary function of leaning towards spam hit rate. People are going to write rules USING this function.) -- Daniel Quinlan anti-spam (SpamAssassin), Linux, http://www.pathname.com/~quinlan/ and open source consulting
