Re: Can not get a filter recognised for one email source

On Wednesday 16 February 2011 10:35:40 you wrote:
> > From: Anne Wilson <[email protected]>
> > To: [email protected]
> > Subject: Can not get a filter recognised for one email source
> > Date: Wed, 16 Feb 2011 09:44:55 +0000
> > 
> > 
> > I get regular emails from one list, that always end up in my Inbox, no
> > matter what I put in the recipes.  At the moment these two recipes
> > 
> > are in place:
> > :0
> > 
> > * (^From|^To|^Subject):
> > .*(Janome|embroidery|Embroidery|emblibrary|empressmills)
> > ${MAILDIR}.INBOX.Sewing/
> > 
> > :0
> > 
> > * ^From: .*Debra\@emblibrary.com
> > ${MAILDIR}.INBOX.Sewing/
> > 
> > Among the headers of the mail are the following lines:
> > 
> > Received: from mail.emblibrary.com ([69.54.45.82]) (envelope-from
> > <[email protected]>) from: [email protected]
> > subject: Feather Your Nest - New at Embroidery Library!
> > X-Envelope-From: [email protected]
> > 
> > none of these trigger a match, and I've no idea why.  Any ideas?
> 
> This is just a *GUESS*.
> 
But it looks to be very likely :-)

>   Background:
>     1) The mail you show is -not- standards compliant.  Message "header"
>        lines are required to start with an upper-case letter.
> 
None of the important-for-this-purpose lines do :-(

>     2) 'Default' for a procmail recipe is to apply the checks against
>        the headers only.
> 
>     3) message format is:
> 	 headers
> 	 blank line
> 	 body
> 
> _Apparently_, when procmail scans the message, and sees something that is
> "not a valid header" _before_ the blank line, it does *NOT* treat it it as
> a header line.
> 
> Two possibilities why:
>   1) it simply 'ignores' that invalid line, and continues on -- looking
>      for the next 'valid' header, or the 'end of headers' blank line.

This appears to be the case.

>   2) it 'assumes' that the 'end of headers' blank line was missed, and that
>      this is 'body' text.
> 
Looking at the source, the blank line appears to be present.

> 
> The 'right' solution to this problem is to get the mailing list to fix
> their BROKEN, NON-STANDARDS-COMPLIANT, software.
> 
:-D  You know how likely that is :-)

> A possible work-around -- *if* 'possibility 2', above applies -- is to
> stick "HB" flags on those two recipes.   Doing that would mean possible
> 'false positives' on other mail, but given the nature of the specific
> patterns, it's very unlikely to be a problem.
> 
> 
> _IF_ 'possibility 1' applies, _and_ the list owner won't fix =their=
> problem, one can repair things on the receiving end, but it takes some
> actual "programming".  Need a program that will convert 'from:' and
> 'subject:', when encountered at the beginning of a line, and before the
> first blank line, to 'From:' and 'Subject:', respectively.  Then, you
> invoke this program, as a 'filter', in a recipe with no conditions,
> _before_ the two recipes you show above.  A crude, ugly, brute-force
> solution, but it will get the job done.

I guess this is the correct answer for this case, but I fear it would be 
beyond my capabilities.  Maybe some of my friends could give me a clue.

Thanks for the speedy answer.  I've been fighting this one for a very long 
time, and at least I know now that it's not being caused by errors on my part.

Anne
-- 
New to KDE Software? - get help from http://userbase.kde.org

____________________________________________________________
procmail mailing list   Procmail homepage: http://www.procmail.org/
[email protected]
http://mailman.rwth-aachen.de/mailman/listinfo/procmail

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.16 (GNU/Linux)

iEYEABECAAYFAk1brnIACgkQj93fyh4cnBf3QwCfdfub8ePnMIdSDKz3CDV8+jbn
pQUAn0XXav5RUdX7I6qla3z6XfDG+nZm
=Sl0k
-----END PGP SIGNATURE-----