[Evolution] Filter question



Hi List,

I'm looking at dspam as an alternative to SpamAssassin, for a couple of
reasons. I won't start any discussion, but both have their own
advantages. Both are MTA-based in my set-ups, so Evo plays no part in
spam determination.

With dspam (pure Bayes) messages come in marked and judged with a
6-decimal probability of being spam. There's also one X-header I want
Evo to filter on: X-DSPAM-Result:. Values are either "spam" or
"innocent". Evo 1.4.5 doesn't seem to want to filter on this header, so
I made a tiny script that looks for a combination of the above on one
line and exits with 0 or 1. I test it at the command line with
script-name message.eml (i.e. script-name $1) and it works, exits with 0
or 1. But it doesn't work as an Evo filter. I probably don't understand
Evo filters - can anyone help me on my way with this? This is mi
scriptlet:

if (( `grep X-DSPAM-Result: $1 | awk '{ print $2 }'`  == "Spam" )); then
  exit 0; else exit 1
fi

This is what the headers look like:

X-DSPAM-Result:  Innocent
X-DSPAM-Result:  Spam

Incidentally, just about every poster on this list is "innocent". Except
Michael. dspam thinks that every mail he posts is spam. I'm busy trying
to teach it that Michaels posts are innocent. It's taking a long time ;)

Thanks!

--Tonni

-- 

We make out of the quarrel with others rhetoric
but out of the quarrel with ourselves, poetry.

mail: billy - at - billy.demon.nl
http://www.billy.demon.nl




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]