[prev in list] [next in list] [prev in thread] [next in thread] 

List:       procmail
Subject:    Re: Adjusting FROM_DAEMON
From:       mark david mcCreary <mdm () internet-tools ! com>
Date:       2009-04-22 0:53:02
[Download RAW message or body]

>>
>
> Actually, what the OP said seems to be the case.  Let me back
> up by saying I'm not much enamored of his recipe.  I'll quickly
> state a couple of things about it below.  But basically,
> I took a sample message I had sitting around, ran it
> through
>


Dallman

Thanks for your feedback.  And I cannot take credit for the recipe, as  
it's from Smartlist in the rc.submit routine.

My version may be a bit old, and I may have tweaked it a number of  
years ago.  But complicated regexp's are not in my skill set.  Here's  
the current version from Smartlist


#
# The following recipe makes sure that:
#       The mail has a sane size (i.e. it is not inordinately big)
#       It does not look like an administrative request.
#       It wasn't sent by this list itself.
#       It wasn't sent by a daemon (misdirected bounce message perhaps).
#

:0
* < $size_limit
* !$^($X_COMMAND:|X-Loop: $\listaddr)
* ! B ?? $^^$X_COMMAND:
* $$daemon_bias
* -100^0 ^FROM_MAILER|\
          ^(((Resent-)?(From|Sender)|X-Envelope-From):|>?From )\
           ([^>]*[^(.%@a-z0-9])?(\
            LIST(SERV|proc)|NETSERV|bounce|autoanswer|echo|mirror\
           )(([^).!:a-z0-9][-_a-z0-9]*)?[%@>      ][^<)]*(\(.*\).*)?)?$ 
([^>]|$)
* 1^0
{

:0
* B ?? >640
* 9876543210^0
{ }




The documentation says FROM_MAILER is a subset of FROM_DAEMON.




>  :0
>  * ^FROM_DAEMON
>  { LOG = "Kilroy was here
> " }
>
> and tested that.  Of course, it did not descend to the action brace- 
> set.
>
> Then I edited the From: address in the message and added "-mail"  
> before
> the "@" sign and did the test again.  It does match.
>
> That aside, some things I don't like about the OP's recipe is
> its unnecessary complexity, the fact that there are presumptive
> tabs that are actually only multiple spaces,[1] and the use of both
> scoring and the empty brace set when one or the other would suffice.
>
> Here's what I'd probably do:
>
>   :0 : named.lock
>   * ^FROM_DAEMON
>   * !^From:.*-mail@
>   | action_line >> whatever




I really like your idea of negating problem email addresses, and am  
going to think about reworking that part of Smartlist.

In addition to -mail I could negate the admin(istrator) type stuff  
too, and add additional exceptions to them as they come up.

And I guess I need to remove the Friendly Name from the From: header  
before comparing it to FROM_DAEMON so that parenthesis do not trip me  
up in the future.

Thanks again for all of your help !

mark

















____________________________________________________________
procmail mailing list   Procmail homepage: http://www.procmail.org/
procmail@lists.RWTH-Aachen.DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail
[prev in list] [next in list] [prev in thread] [next in thread]