On Wed, 2009-05-20 at 00:20 +0200, Jonas Eckerman wrote: > Jari Fredriksson wrote: > > > As the mail contains no text, there propably is not much to learn. > > Why not? Bayes learns from headers as well, and headers can be just as > useful as body text for classifying mail.
Indeed. Hence my insisting to feed them to sa-learn. Also, there actually is text. The Subject is part of the rendered "body" text parts, thus used as regular text tokens by Bayes. See M::SA::Message.pm and M::SA::Bayes.pm respectively. guenther -- char *t="\10pse\0r\0dtu...@ghno\x4e\xc8\x79\xf4\xab\x51\x8a\x10\xf4\xf4\xc4"; main(){ char h,m=h=*t++,*x=t+2*h,c,i,l=*x,s=0; for (i=0;i<l;i++){ i%8? c<<=1: (c=*++x); c&128 && (s+=h); if (!(h>>=1)||!t[s+h]){ putchar(t[s]);h=m;s=0; }}}