[26419] in Athena Bugs
sun4 9.3.18: send
daemon@ATHENA.MIT.EDU (Jacob Morzinski)
Tue Apr 26 12:37:45 2005
Message-Id: <200504252214.j3PMEdow022314@kamp-krusty.mit.edu>
To: bugs@mit.edu
Date: Mon, 25 Apr 2005 18:14:37 -0400
From: Jacob Morzinski <jmorzins@mit.edu>
X-Spam-Score: -4.9
X-Spam-Flag: NO
Errors-To: bugs-bounces@mit.edu
System name: kamp-krusty.mit.edu
Type and version: Sun-Blade-100 9.3.18 (with mkserv)
Display type: ifb
Shell: /bin/athena/bash
Window manager: sawfish
What were you trying to do?
I'm using MH-E 7.82, an emacs interface to [n]mh.
Under certain conditions, MH-E calls "send -mime"
to send a user's mail. "send -mime" fails.
(Yes, I do realize this is a low-priority bug.
No action is needed now, but you may have to decide
what to do about this when future problems come up.)
What's wrong:
Although the nmh man pages claim that "send -mime" ought
to work, and "send -help" claims that -mime should be a valid
command-line flag, using it results in an error message:
kamp-krusty{107}$ send -help | grep mime
-[no]mime
kamp-krusty{108}$ comp
Waiting for Emacs...Done
What now? send -mime
spost: -mime unknown
send: message not delivered to anyone
What now?
What should have happened:
My ideal solution would be for "send -mime" to silently
ignore the "-mime" (or "-nomime") flag.
I believe that Athena has a local patch to MH, put in
to prevent BCC'd messages from being mangled in the
traditional way that MH mangles them. (Standard MH will
reformat a message, adding body text and removing some
headers.)
The purpose of the "-mime" flag is to enable this BCC
reformatting to be done in a mime-friendly fashion.
(Without the "-mime" flag, the BCC reformatting would
corrupt mime-encoded messages.)
I don't like the message reformatting, and I very much
appreciate the Athena patch that disables it. Since
Athena MH does not reformat BCC messages, it has no need
for the -mime flag. However, some software tries to use
the -mime flag. We could change each program that tries
to use "send -mime", or we could change "send".
In the short term, I can patch the sipb copy of MH-E 7.82
so that it does not try to use "send -mime", and you
don't have to do anything. I'm sending this report as
a heads-up message, to let you know that you'll have to
decide what to do about the "-mime" flag in the future,
for programs that expect nmh to have such a flag.
Please describe any relevant documentation references:
Here's an excerpt from send(1), for convenience:
If a "Bcc:" field is encountered, its addresses will be used
for delivery, and the "Bcc:" field will be removed from the
message sent to sighted recipients. The blind recipients
will receive an entirely new message with a minimal set of
headers. Included in the body of the message will be a copy
of the message sent to the sighted recipients. If
`-filter filterfile' is specified, then this copy is fil-
tered (re-formatted) by mhl prior to being sent to the blind
recipients. Alternately, if you specify the `-mime' switch,
then send will use the MIME rules for encapsulation.
Thanks for your time,
Jacob Morzinski
