Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxn!ihnp4!qantel!lll-lcc!lll-crg!seismo!brl-sem!brl-smoke!gwyn
From: gwyn@brl-smoke.ARPA (Doug Gwyn )
Newsgroups: net.unix-wizards
Subject: Re:  Manuals, windows, auto-newlines
Message-ID: <238@brl-smoke.ARPA>
Date: Wed, 23-Apr-86 18:28:41 EDT
Article-I.D.: brl-smok.238
Posted: Wed Apr 23 18:28:41 1986
Date-Received: Tue, 29-Apr-86 00:42:09 EDT
References: <2653@brl-smoke.ARPA> <1098@whuxl.UUCP> <137@brl-smoke.ARPA> <1101@whuxl.UUCP>
Reply-To: gwyn@brl.ARPA
Organization: Ballistic Research Lab (BRL)
Lines: 25

In article <1101@whuxl.UUCP> mike@whuxl.UUCP (BALDWIN) writes:
>True; it is better that rows/columns be in the terminal
>ioctl structure (termio/sgttyb).  Has 4.3 also moved $TERM
>there?  TERM (and LINES and COLUMNS) are per-port variables,
>not per-process.

No $TERM in the kernel, but using shell functions I can solve this
(so long as the functions can be passed in the environment a la
8th Edition UNIX; otherwise one needs to source a "function
definition file" when a shell starts up).  Just make the command
that downloads a terminal emulator a shell function that first
adjusts $TERM.

>Then there's the problem of cu'ing from a window.  Not much
>can be done in that case if you reshape.  What happens with
>remote shells?  Do they see the change or SIGWINCH?

"cu", "rlogin", etc. just have the problem that not enough
terminal state/control information is passed in their protocols.
With stream i/o networking, and suitable protocols, it should
be possible to have the "terminal handler" on the remote end
completely aware of the terminal state.  Occasionally we even
get that here using 4.2+BSD "rlogin", so far as initial window
size is concerned.  (I don't think window changes are passed
along, but they could be and should be.)