[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Nmh-workers] nmh 1.1 annotate
From: |
Jerry Peek |
Subject: |
Re: [Nmh-workers] nmh 1.1 annotate |
Date: |
Thu, 01 Jul 2004 14:08:25 -0700 |
On 1 July 2004 at 14:35, Vish Parakala <address@hidden> wrote:
> Yup, anno seems to work perfectly fine ...
> it's just all other components (which use post/send)
About 15 years ago, when I started writing the O'Reilly MH book,
I played around a lot with the mostly-undocumented process
environments under MH: how certain environment variables passed
from a parent process to its children would alter the behavior
of some back-end applications. For instance, if I remember
right, dist sets an environment variable to tell "send" or "post"
to at the "Resent-to:" header field instead of "To:", and so on.
(Most of that stuff is documented in the online version of the
book at http://www.ics.uci.edu/~mh/book/ . Try an index search
for "environment". I haven't completely updated this for nmh.)
One quick way to track down the problem might be to compare an
older nmh installation with a newer one. Make a sendproc script,
or a postproc script, like this:
#!/bin/sh
echo "$0 environment:"
env
echo "$0 arguments:"
echo "$*"
exit 1
(You set "sendproc:" -- and, I think, "postproc:" -- from the
.mh_profile file.) Send similar messages from both systems
and see if something is different. I have no idea if that will
help; it's just an idea.
Jerry
--
Jerry Peek, address@hidden, http://www.jpeek.com/