procmail
[Top] [All Lists]

problem with formail splitting

2000-11-25 12:26:21
(Perhaps this belongs on the procmail-dev list; feel free to
redirect there if appropriate.)

I observed "weird" behavior when trying to split a mailbox, and
have traced it down to a sometimes-inserted header which is formatted
something like "MBOX-Line: {stuff that looks like From_}"

It seems as if this comes -- sometimes, not always -- with
messages forwarded by yahoo.com.  It also seems to prevent
formail -s from recognizing the start of a message in some cases.
What's up?

Note also, below, that changing it to X-MBOX-Line: makes it
work with formail.  Is formail doing something tricky by trying
to analyze headerfield names?  The man page says nothing about this
that I can see.

Here are specific test cases with a simplified input
(see "cat datafile" at the end); I've indented everything by
one space just to avoid confusion about From_ handling that
might occur when this list mail is delivered:
=====
 % cat datafile | formail -s head -1
 From first(_at_)sender Fri Nov 17 20:41:23 2000
 %
=====
 % egrep -v MBOX datafile | formail -s head -1
 From first(_at_)sender Fri Nov 17 20:41:23 2000
 From second(_at_)sender Thu Nov 16 20:41:23 2000
 %
=====
 % sed -e s/MBOX/X-MBOX/ datafile | formail -s head -1
 From first(_at_)sender Fri Nov 17 20:41:23 2000
 From second(_at_)sender Thu Nov 16 20:41:23 2000
 %
=====
 % cat datafile
 From first(_at_)sender Fri Nov 17 20:41:23 2000
 MBOX-Line: From first(_at_)sender Fri Nov 17 11:28:20 2000
 Date: Fri, 17 Nov 2000 02:25:48 -0500

 first body

 From second(_at_)sender Thu Nov 16 20:41:23 2000
 MBOX-Line: From second(_at_)sender Thu Nov 16 11:28:20 2000
 Date: Thu, 16 Nov 2000 02:25:48 -0500

 second body

 %
=====

Thanks,
Stan Ryckman

ps - the "useless cat" above is for symmetry with the other test cases.
:-)
_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail

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