undoing mailscanner modifications before feeding into sa-learn

[Wolfgang Kohnen (FEYERABEND)]

    [ The following text is in the "ISO-8859-15" character set. ]
    [ Your display is set for the "US-ASCII" character set.  ]
    [ Some characters may be displayed incorrectly. ]

Hi folks,

I want to feed back spam and ham messages to the bayes filter.  I've
setup two shared folders (ham & spam) on my imap machine and a fetchmail
which fetches the messages from there and puts it into sa-learn with
fetchmails --mda feature.  This works fine so far.

The docs of sa-learn say, that it internally does some undos (like
spamassassin -d) in order to remove mail headers which originates by
spamassassin so that there is no bias reinforcement.

My question:  is sa-learn aware of how mailscanner modifies the message
and can sa-learn undo it?  How do I have to tweak mailscanner?  For
example, I will have to turn off disarming of html and other destructive
conversions.  Any experiences?  The faq-o-matic doesn't note this issue
at all.

TIA,
Wollie

------------------------ MailScanner list ------------------------
To unsubscribe, email jiscmail at jiscmail.ac.uk with the words:
'leave mailscanner' in the body of the email.
Before posting, read the MAQ (http://www.mailscanner.biz/maq/) and
the archives (http://www.jiscmail.ac.uk/lists/mailscanner.html).

Support MailScanner development - buy the book off the website!