emacs -nw keypad

Ken Brown kbrown@cornell.edu
Wed May 27 17:01:00 GMT 2009

On 5/26/2009 9:11 PM, Tim Adye wrote:
> In Cygwin 1.5 xterm (TERM=xterm), with "emacs -q -nw" (or also with "-f
> tpu-edt") I get
> ESC [ > 1 ; 2 4 2 ; 0 c ESC O q ESC O r ESC O s ESC
> O t ESC O u ESC O v ESC O P l
> The initial "ESC [ > 1 ; 2 4 2 ; 0 c" is just the response from xterm, asked
> for its version number (242). The "ESC O q" to "ESC O v" are the keypad
> "123456", but somehow with numlock on (perhaps this is a feature of my
> Exceed X-server). The "ESC O P l" is the "F1 l" at the end.
> Despite the differences, it looks like neither of us is getting any
> interpretation of the keys.

As I understand the emacs documentation, the setting TERM=xterm should 
cause emacs to load term/xterm.el.  In that file I find lines like

(define-key map "\eOq" [kp-1])
(define-key map "\eOr" [kp-2])

This looks like the place where emacs should learn to interpret the 
keypad keys.  So is this library failing to load for some reason?  Is 
there some initialization that emacs does (perhaps using the ncurses 
library) that overrides the setting of TERM?  I mentioned ncurses 
because that's something else that has been updated recently in the 
cygwin distribution.


P.S.  I said in an earlier email that, after starting emacs, TERM has 
the value "dumb".  This turns out to be irrelevant.  Emacs apparently 
does this for the benefit of subprocesses that it might start.

Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

More information about the Cygwin mailing list