X-Spam-Headers
Adrian Otto
aotto at aotto.com
Wed Sep 25 01:41:01 CEST 2002
David,
Please commit your patch.
Thanks,
Adrian
> -----Original Message-----
> From: David Relson [mailto:relson at osagesoftware.com]
> Sent: Monday, September 23, 2002 8:27 PM
> To: bogofilter-dev
> Cc: Adrian Otto; Matthew E. Martini
> Subject: X-Spam-Headers
>
>
> At 08:53 AM 9/12/02, Matt Martini wrote:
> >-----BEGIN PGP SIGNED MESSAGE-----
> >
> >On Tue, 10 Sep 2002, Adrian Otto wrote:
> >
> > >
> > > ... [snip] ...
> > >
> > > 2) If a message already has an X-Spam-Status header, it is
> ignored instead
> > > of passed through in -p mode. A syslog message is logged if
> an existing
> > > header is skipped.
> >
> >I would also suggest ignoring the foolsing headers:
> >
> > X-Spam-JunkFilter
> > X-jf
> >
> > X-Spam-Status
> > X-Spam-Flag
> > X-Spam-Level
> > X-Spam-Checker-Version
> > X-Spam-Report
>
> The processing of passthrough in main.c has been optimized. Part of this
> optimization was to remove the check for X-Spam-Status header
> lines. However, if the message already has such lines, the
> original lines
> will be left in it _and_ the new line will be added. The
> original idea for
> this optimization included having get_token() check for, and remove, the
> unwanted lines. An alternate approach would be to have the
> grammar remove
> these lines. It appears that doing this is as simple as:
>
> diff -u -r1.7 lexer_l.l
> --- lexer_l.l 23 Sep 2002 11:31:53 -0000 1.7
> +++ lexer_l.l 24 Sep 2002 03:25:14 -0000
> @@ -273,6 +273,7 @@
> ^From\ {return(FROM);}
> ^Date:.*|Delivery-Date:.* ;
> ^Message-ID:.* ;
> +^X-Spam-[A-Za-z]+:.* ;
> ^{BASE64}+$ ;
> ^\tid\ .* ;
>
>
> Any objections to my committing this patch?
>
> David
>
>
>
>
>
More information about the bogofilter-dev
mailing list