multibyte UTF-8 in quoted-printable and base64

I'm running dspam 3.9.0 built from the FreeBSD port, and it seems to
be struggling with multi-byte UTF-8 in quoted-printable and base64 mime
sections.

I get some spam in UTF-8 Greek and, since I don't receive any legitimate
mail in Greek, it should be caught easily. What I'm seeing is that
most aren't caught, and those that are are either caught by headers
only or have Greek body tokens from an 8bit mime section. 

I am seeing some English tokens from encoded UTF-8 in X-DSPAM-Factors,
but not the multi-byte Greek. The same thing happens in html
and plain text.


------------------------------------------------------------------------------
Comprehensive Server Monitoring with Site24x7.
Monitor 10 servers for $9/Month.
Get alerted through email, SMS, voice calls or mobile push notifications.
Take corrective actions from your mobile device.
http://p.sf.net/sfu/Zoho