> > That's how you can investigate the Bayes tokens for the messages that
> > score neutral, despite learning. Isn't that what you asked for?
>
> Something like that. I interpreted that as to expand the headers with a more
> verbose line. I just checked a recently treated (and cleared) incoming
> header, and the line is added, but its otherwise empty. So is the sa status
> box kmail gives me. Duh.
Hmm, did you --lint your changes?
Rather than immediately applying this to all incoming mail, I'd try this
with the offenders only. That is, run them through 'spamassassin'
manually, adding the specific options using --cf. Maybe adding them to
local.cf temporarily, without restarting spamd.
That way, you can specifically investigate the under-performers.
> But I'd expect to see some details there if its a 4 star message.
The overall score of the message is unrelated to the Bayes tokens, other
than getting a score for Bayes rules. But you're probably right that the
header should hold at least some information.
--
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; }}}
