Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
Sender: cygwin-owner AT sourceware DOT cygnus DOT com
Message-Id: <3.0.5.32.19990215095816.009399c0@london.beasys.com>
Date: Mon, 15 Feb 1999 09:58:16 +0000
To: "David Starks-Browning" <starksb AT ebi DOT ac DOT uk>, cygwin AT sourceware DOT cygnus DOT com,
        xemacs-nt AT xemacs DOT org
From: Andy Piper <andy AT xemacs DOT org>
Subject: Re: cygwin build of xemacs (-nw) and inetd/telnetd
In-Reply-To: <14022.65333.423000.845970@gargle.gargle.HOWL>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

At 04:52 PM 2/14/99 +0000, David Starks-Browning wrote:
>I just completed a cygwin B20.1 build of xemacs-21.0.63 on NT4/SP4.  I
>am also using Sergey's remote package, providing inetd & telnetd.  The
>goal of this effort is to run xemacs -nw in a (remote) telnet session.
>GNU nt-emacs is known not to work under these circumstances.
>
>Unfortunately, xemacs does not seem to be getting certain keystrokes,
>namely DEL and C-g.  Is there something special that I must do for
>these to work properly?

There is some wierd voodoo to do with DEL that got introduced in 21.0,
someone may be able to answer this question on xemacs-beta (to which I have
copied). ^G not working seems wierd to me. There was a bug in cygwin tty
handling which meant ^G got interpreted wrongly - but this was fixed ages
ago (back in b19). You could try generating an strace and see what comes
out when you hit ^G.

>I have confirmed that info 2.18 (from GNU texinfo 3.12) sees ^G and
>^? fine, by inspecting a dribble file.  This should rule out Sergey's
>remote package (or my telnet client app) as the culprit.  On the other
>hand, none of this is a problem when xemacs -nw is invoked in a local,
>"cygnus.bat" shell with CMD.EXE.
>
>No doubt there is much I don't understand about xemacs keys &
>bindings.  In case it's relevant, I found this in the PROBLEMS file:
>
>> *** You type Control-H (Backspace) expecting to delete characters.
>> 
>> Emacs has traditionally used Control-H for help; unfortunately this
>> interferes with its use as Backspace on TTY's.  One way to solve this
>> problem is to put this in your .emacs:
>> 
>>   (when (eq tty-erase-char ?\C-h)
>>     (keyboard-translate ?\C-h ?\C-?)
>>     (global-set-key "\M-?" 'help-command))
>> 
>> This checks whether the TTY erase char is C-h, and if it is, makes
>> Control-H (Backspace) work sensibly, and moves help to Meta-? (ESC ?).
>
>It almost seems like this is already implemented:  after "stty erase ^H", 
>C-h generates "backspace", and after "stty erase ^?", C-h generates
>help-command.  (Sorry, I probably have the terminology wrong.)
>In both cases, neither ^? (\177) nor ^G (\007) seem to register with
>xemacs.
>
>Cygwin bash is my login shell per /etc/passwd, TERM=linux and
>CYGWIN="tty binmode".  (~/.bash_profile execs a new bash after setting
>CYGWIN.)  I've built xemacs with and without ncurses (4.12) with no
>effect.

andy