procmail
[Top] [All Lists]

Re: Fw: despamassassing

2004-11-17 16:41:21
On Wed, Nov 17, 2004 at 11:35:49PM +0100, Hanspeter Roth wrote:
  On Nov 17 at 19:31, Dallman Ross spoke:

If he wants to remove the headers, he can read "man formail".

  :0 fw hi
  * ^X-Spam-
  | formail -I X-Spam-

This probably adds some X-Spam- headers. But that doesn't help me.

Perhaps you might wish to *read* what I typed, HR.  "If he
wants to remove the headers . . ."  Now, why would you think
from that that I told you how to *add* some headers?
(And typing "man formail" would not be so difficult, would it?)


SpamAssassin puts the original message into an attachment and puts
it's own comments into the main message. This makes it rather
annoying to read the original message by mutt. Also the coloring of
mutt is lost. And further the headers are now moved to the body
which makes it less efficient to filter by procmail.

Since what SP does is configurable based on who ran it in
the first place and since SP does different things depending
on what version the person running it is using it, I think
you might have a mess on your hands trying to undo all those
things without SA and without perl.  But in any case, though
there might be a way to strip out all that other stuff, I don't
know that it's a procmail issue.  Maybe you should ask on the
SA list?

Otoh, if the original message really is inside, you conceivably
*could* figure out some incredibly hoary way to do it in procmail.
It seems like the wrong tool, however.

One tool that could be useful here is reformime, which you
would have to grab from wherever its source is and compile.
I got it installed on our system a couple of years ago, and it's
pretty nifty.  It can identify mime subparts and extract
them based on some criteria.

-- 
dman

____________________________________________________________
procmail mailing list   Procmail homepage: http://www.procmail.org/
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail

<Prev in Thread] Current Thread [Next in Thread>