[Spambayes] RE: Ugh...imapfilter woes continue

[Tony Meyer]

> Well, I got the MySQL database working after some helpful 
> tips from Tony, but immediately ran into a problem. Thinking 
> that it may have been MySQL vs. Spambayes, I changed to using 
> a pickle, and ran into the same problem:

Was it definitely the same problem?  i.e. the error message was exactly the
same?  That indicates that imapfilter is doing something wrong.

[...]
>     assert spamcount <= nspam, "Token seen in more spam than 
> spam trained."
> AssertionError: Token seen in more spam than spam trained.

It looks like you're running from CVS, so I'm picking that imapfilter is at
fault here.  Imapfilter has gone through fairly substantial changes in CVS
(compared to 1.0), so it's quite possible that a bug has been introduced.
Unfortunately, I'm not sure how this particular problem could be caused -
the only thing I can think of is that training has been interrupted at some
point - has imapfilter crashed during training?

> I remember seeing something in an email a little while back 
> about fixing this error by exporting the db into a cvs and 
> correcting it manually, but I can't find any details on what 
> exactly needs to be done. And seeing as I've run into this 
> exact same error twice now, immediately after training, is 
> there something I am doing when training that causes this error.

You can do that, but it's better to just start from scratch.  It involves
guesswork as to which numbers to use, so you're putting the reliability of
the database into question.  A message with the steps is here:

<http://mail.python.org/pipermail/spambayes/2004-October/014902.html>

=Tony Meyer

---
Please always include the list (spambayes at python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes. This
way, you get everyone's help, and avoid a lack of replies when I'm busy.