rmail and attachments (emacs 23)

[Art Werschulz]

Hi all.

Most of us are using the emacs rmail system for reading (and sending)
mail. 

Previously, we were using emacs 22 on our Fedora 10 systems.  We used
John Rulnick's "etach" package to handle attachments (either detaching
them when reading mail, or attaching them when sending mail).

We have now upgraded from Fedora 10 to Fedora 11, which induced an
upgrade from emacs 22 to emacs 23 (RPM emacs-23.1-3.fc11.i586, to be
precise).  

etach still works fine for attaching attachments to mail that I'm
sending.  However, etach does not work for detaching attachments from
mail that I'm reading.

I have tried using rmail's rmail-mime command, bound to the key "v", to
deal with rmail msgs containing attachments.  Sometimes this works;
sometimes this doesn't.  When it fails, I get the following error msg:

Debugger entered--Lisp error: (error "Malformed boundary; type: 
(multipart/alternative (boundary . =_alternative 0075D08205257633_=)); 
disposition: (inline); encoding: nil")
  signal(error ("Malformed boundary; type: (multipart/alternative (boundary . 
=_alternative 0075D08205257633_=)); disposition: (inline); encoding: nil"))
  error("%s; type: %s; disposition: %s; encoding: %s" "Malformed boundary" 
("multipart/alternative" (boundary . "=_alternative 0075D08205257633_=")) 
("inline") nil)
  rmail-mm-get-boundary-error-message("Malformed boundary" 
("multipart/alternative" (boundary . "=_alternative 0075D08205257633_=")) 
("inline") nil)
  rmail-mime-multipart-handler(("multipart/alternative" (boundary . 
"=_alternative 0075D08205257633_=")) ("inline") nil)
  rmail-mime-handle(("multipart/alternative" (boundary . "=_alternative 
0075D08205257633_=")) ("inline") nil)
  rmail-mime-show()
  rmail-mime-multipart-handler(("multipart/mixed" (boundary . "=_mixed 
0075D08005257633_=")) ("inline") nil)
  rmail-mime-handle(("multipart/mixed" (boundary . "=_mixed 
0075D08005257633_=")) ("inline") nil)
  rmail-mime-show(t)
  rmail-mime()
  call-interactively(rmail-mime nil nil)

Again, sometimes the "v" command works fine; sometimes, it doesn't.
I've been unable to figure out necessary and sufficient conditions for
success or failure.

Any idea why this is happening?
Any idea how to fix the problem?

Thanks.

-- 
Art Werschulz (8-{)}   "Metaphors be with you."  -- bumper sticker
GCS/M (GAT): d? -p+ c++ l++ u+ P++ e--- m* s n+ h f g+ w+ t+ r- 
Net: agw@dsm.fordham.edu http://www.dsm.fordham.edu/~agw
Phone:   Fordham U. (212) 636-6325, Columbia U. (646) 775-6035