My system uses qmail-scanner to check incoming mail for viruses, and in the same run SpamAssassin evaluates the email. Apparently the only indication the
user (and hence TMDA) gets is a header line that is split into three lines similar to the following example:
Received: from [EMAIL PROTECTED] by blazingangles.com by uid 0
with qmail-scanner-1.20
(clamscan: 0.68. spamassassin: 2.63. Clear:RC:0(195.41.53.68):SA:1(7.8/5.0):.
Processed in 7.523049 secs); 01 Apr 2004 12:24:57 -0000
(clamscan: 0.68. spamassassin: 2.63. Clear:RC:0(195.41.53.68):SA:1(7.8/5.0):.
Processed in 7.523049 secs); 01 Apr 2004 12:24:57 -0000
That ought to be
enough, though, because the ":SA:1:(7.8/5.0):" part indicates that this
particular email scored 7.8 SpamAssassin points.
I want to catch mail that
scores 5.0 or above according to SpamAssassin. I've tried to accomplish this
with the following line in my incoming filter:
headers 'spamassassin:
.*:SA:[0|1]([4-9]\.[0-9]/[0-9]\.[0-9])' confirm
I've verified
that the above regular _expression_ works with the unix 'grep' command, so
unless there are some Perl-ish peculiarities involved, the above line
should catch any email scoring SpamAssassin verdicts between 5.0 and
9.9. (Strictly speaking, the "spamassassin: .*" part shouldn't be necessary, but
it shouldn't hurt either.)
TMDA never seems to
locate any such header content, however. What am I doing wrong? Or could it be
that maybe TMDA doesn't process
multi-line headers?
- wolf
-
_____________________________________________ tmda-users mailing list ([EMAIL PROTECTED]) http://tmda.net/lists/listinfo/tmda-users
