[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
- Subject: spamass-milter bug
- From: dnelson at allantgroup.com (Dan Nelson)
- Date: Wed Nov 24 07:45:24 2004
In the last episode (Nov 24), Vince Fleming said:
> There seems to be some conflicting info on the 'net about who's
> maintining spamass-milter - both of these email addresses are
I'm the maintainer of the software, but not the port itself. I should
file a PR assigning that to me, I guess :)
> Anyway - I found a small bug.
> The -r command-line option doesn't work with SpamAssassin v3.
> Here's why:
> reject_score is an 'int', and should be a 'float', as the scores are
> not integers: ie: 5.2 or -1.1
All that means is you can't ask to reject at 5.2; you can only set it
to 5 or 6. How often to you really want to reject on a fractional
> Also, in function assassinate(), getting the score from
> assassin->spam_status via sscanf() should look for "score=%f", not
Same thing; sscanf will parse "score=1.5" using "%*s score=%d" and
assign score the value 1. The port commit on 2004/10/15 changed the
sscanf string to patch SA 3.0's output, so make sure your portversion
is at 0.2.0_5 and you should be set.
The CVS version of spamass-milter checks for both hits= and score=.
Visit your host, monkey.org