🚀 go-pugleaf

In article <1992Dec15.154511.27086@cs.hw.ac.uk> hmc@cs.hw.ac.uk (Hugh Conner) writes:
>We have several 486 based systems running ESIX 4.0.4. When using the mailx
>command to read mail messages a problem occurs if the first word on a line
>of the message is 'from'. I notice that when sending mail if the first word
>is 'From' with a capital F it replaces it with '>From' which I assume is meant
>to stop the mail reader treating it as the start of a new message. This does
>happen with 'from' and although /bin/mail reads the message ok, mailx
>decides that the line starting 'from' is a new message. Does anyone know a
>patch for this or should I just compile a decent mail reader and scrap mailx.

Pre-sysvr4 mail used lines starting with "From " as the delimiters between
messages in the mailbox file, and added the leading ">" to any such
lines in the body of a message.  SysVr4 mail uses a Content-Length: header
to encapsulate the body and does not restrict or modify the content
of the body.
I'd recommend getting the current version of ELM which understands this
convention and is a lot nicer than mailx anyway.

Les Mikesell
 les@chinet.chi.il.us

In article <1992Dec15.154511.27086@cs.hw.ac.uk>, hmc@cs.hw.ac.uk (Hugh Conner)
writes:
|> We have several 486 based systems running ESIX 4.0.4. When using the mailx
|> command to read mail messages a problem occurs if the first word on a line
|> of the message is 'from'. I notice that when sending mail if the first
|> word
|> is 'From' with a capital F it replaces it with '>From' which I assume is
|> meant
|> to stop the mail reader treating it as the start of a new message. This
|> does
|> happen with 'from' and although /bin/mail reads the message ok, mailx
|> decides that the line starting 'from' is a new message. Does anyone know a
|> patch for this or should I just compile a decent mail reader and scrap
|> mailx.

This is a known problem in mailx. Individual messages are supposed to be
divided
by lines starting with "From ", not "from" or "from ".  But mailx is doing
this without
looking at the case. You need a fix from your vendor. Or a decent mailreader
like
ELM. :-))

	Regards
		Bo

-- 
Bo Asbjorn Muldbak             // Email:  Bo.Muldbak@ColumbiaSC.NCR.COM
Unix Systems Business Unit     // Phone:  +1 (803) 791-6871
NCR Corporation, MCPD-Columbia // Fax  :  +1 (803) 739-7371
All standard disclaimers apply // Wizards? Where, when, why? I'm an atheist..

hmc@cs.hw.ac.uk (Hugh Conner) writes:
: We have several 486 based systems running ESIX 4.0.4. When using the mailx
: command to read mail messages a problem occurs if the first word on a line
: of the message is 'from'. I notice that when sending mail if the first word
: is 'From' with a capital F it replaces it with '>From' which I assume is meant
: to stop the mail reader treating it as the start of a new message. This does
: happen with 'from' and although /bin/mail reads the message ok, mailx
: decides that the line starting 'from' is a new message. Does anyone know a
: patch for this or should I just compile a decent mail reader and scrap mailx.

	This indeed is one of the known problems with mailx. There is one more
problem with mailx while reading mail. If any of the mail messages have
non-text byte, then the mailx reading goes berserk. It for the reasons I
fail to fathom, appends the mail portion after that funny character at the
end of the mail box and the effect is that you see your mailbox growing
EXPONENTIALLY. You neither can delete the mail message which has this funny
byte nor delete the mail messages after that as mailx fail to recognize the
mail boubdaries. 

	No easy patch I suppose. We have switched to MH and we are very happy
with it.
 MH source :

  via anonymous ftp:
    FTP Site:		IP Address:	Path/File Name:		Size:
    -----------------------------------------------------------------
    ftp.ics.uci.edu	[128.195.1.1]	mh/mh-6.7.tar.Z		1.6MB
    louie.udel.edu	[128.175.1.3]	portal/mh-6.7.tar.Z	1.6MB
    ftp.uu.net		[137.39.1.9]	mail/mh/tar/mh-6.7.tar.Z-split/	256K*7
  					README, part01, ..., part07

Cheers,
nitin

--
----------------------------------------------------------------------
Nitin Kaulavkar, SysAdm, <nitin@cse.iitb.ernet.in>
S/W and Computing Laboratory, Dept of C.S.E. ,IIT, Bombay 400 076.
----------------------------------------------------------------------

< From: hmc@cs.hw.ac.uk (Hugh Conner)

< We have several 486 based systems running ESIX 4.0.4. When using the mailx
< command to read mail messages a problem occurs if the first word on a line
< of the message is 'from'. I notice that when sending mail if the first
< word is 'From' with a capital F it replaces it with '>From' which I assume
< is meant to stop the mail reader treating it as the start of a new
< message. This does happen with 'from' and although /bin/mail reads the
< message ok, mailx decides that the line starting 'from' is a new message.
< Does anyone know a patch for this or should I just compile a decent mail
< reader and scrap mailx.

This was fixed in System V release 4.1 ES. The fix should be available from
your vendor.

					Tony Hansen
			    hansen@pegasus.att.com, tony@attmail.com
				att!pegasus!hansen, attmail!tony

hansen@pegasus.att.com (Tony L. Hansen) writes:

>< From: hmc@cs.hw.ac.uk (Hugh Conner)

<mailx problem description deleted>

>This was fixed in System V release 4.1 ES. The fix should be available from
>your vendor.

Bah hah hah hah

>					Tony Hansen
>			    hansen@pegasus.att.com, tony@attmail.com
>				att!pegasus!hansen, attmail!tony
-- 
Steven C. Monroe            VOICE:(703)406-2082                PC Three, Inc.
scum@pcthree.com             FAX:(703)406-2078                  P.O. Box 1644
                                                              Herndon, VA 22070

In article <1992Dec28.193828.17739@cbnewsk.cb.att.com>
	hansen@pegasus.att.com (Tony L. Hansen) writes:

>< From: hmc@cs.hw.ac.uk (Hugh Conner)

>< We have several 486 based systems running ESIX 4.0.4. When using the mailx
>< command to read mail messages a problem occurs if the first word on a line
>< of the message is 'from'.

>This was fixed in System V release 4.1 ES. The fix should be available from
>your vendor.

Version 4.1, eh?

A version that nobody has, since everyone is either still at 4.0
(Dell, ESIX) or skipped to 4.2 (Consensys, Univel).

As I recall, a move up a release number from a vendor requires extra $$$
to USL. ESIX told me that part of the reason their 4.0.3 to 4.0.4
license fee is more expensive than they (ESIX) wanted was because of an
upgrade fee to USL.

Would USL offer bug fixes for individual problems in 4.0 without
requiring their licensees to move to 4.1? Most vendors like ESIX, Dell
and SCO have offered update patches at no cost other than the cost of
retrieving them. If only USL was as responsive to *its* customers.

Sigh.

I, for one, will hardly mourn the departure of AT&T from USL's
ownership (and thus, leadership). Novell may not be any better,
but it can't possibly be worse.

-- 
 Evan Leibovitch, Sound Software Ltd., located in beautiful Brampton, Ontario
         evan@telly.on.ca / uunet!utzoo!telly!evan / (416) 452-0504
         "IBM's three letters are probably EBCDIC for CYA" - Nik H.

In article <2B4061C0.3F79@telly.on.ca> evan@telly.on.ca (Evan Leibovitch) writes:
>
> [ ... stuff about support policies deleted ...]
>
>I, for one, will hardly mourn the departure of AT&T from USL's
>ownership (and thus, leadership). Novell may not be any better,
>but it can't possibly be worse.

It can if Novell refuses to re-license all the non-Novell vendors, or
increases the price to the levels associated with Netware.  That would
kill any derivatives of UNIX for all but a few corporate sites, and
would certainly kill any chance of being able to choose among vendors.
I sure hope Novell does't do that - I cannot afford 12 - 18K for a home
UNIX site.  And a 'reasonably' priced version that is crippled to only
3 to 5 logins, and/or by removal of other features (such as the networking
capability, compiler and libs, etc., unless many extra thousands of
dollars are paid) is not something that I would look forward to, either.

That is somnething I have not seen discussed (or heard of anything from
Novell indicating a policy) - will the price for OS software derived from
USL code remain the same, go lower, or be increased by a factor of 6?

Anybody know?  <while I sit here chewing on fingernails...>

-- 
pat@rwing.uucp                                  (Pat Myrto),  Seattle, WA
                      If all else fails, try:
         ...!uunet!{pilchuck, polari}!rwing!pat
WISDOM:    "Travelling unarmed is like boating without a life jacket" 

< From: evan@telly.on.ca (Evan Leibovitch)
<< This was fixed in System V release 4.1 ES. The fix should be available
<< from your vendor.

< Version 4.1, eh?
<
< A version that nobody has, since everyone is either still at 4.0 (Dell,
< ESIX) or skipped to 4.2 (Consensys, Univel).

If you have 4.2, then you have the fixes put into 4.1 as well. Enough said.

					Tony Hansen
			    hansen@pegasus.att.com, tony@attmail.com
				att!pegasus!hansen, attmail!tony