[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Request for help with X keyboard config

On Wed, Apr 01, 1998 at 09:39:06PM +1200, Carey Evans wrote:
> Mark Baker <mbaker@iee.org> writes:
> > For xterm you need 
> > 
> > Xterm*VT100*translations:       #override \
> >         <Key>BackSpace:                 string(0x7F) \n\
> >         <Key>Delete:                    string("ESC[3~")
> That should be XTerm*... and "\033[3~".
>                 ^
> Thanks for explaining what needs to be set though.

I've tried this already, and that's part of what my complaint is.

The <Key>BackSpace bit seems to work, but <Key>Delete simply echoes a tilde
when it is pressed (try it).

In other news, tcsh and bash seem to have completely different expectations
of what key does what when they are launched from within an xterm.

In tcsh, both ^H and ^? do a delete-previous-char, i.e., a destructive
backspace.  The above delete sequence ("\033[3~") simply echoes a tilde.
This behavior is consistent both in VC's and within xterms.

Bash is a more interesting story.  In a VC, ^H, ^?, and \033[3~ all perform
a destructive backspace.  In an xterm, ^H does a destructive backspace, ^?
does nothing but beep (not mapped?), and \033[3~ echoes a tilde, just as
tcsh does.

There are behavioral inconsistencies in mutt and vim as well, depending on
whether they're in an xterm or VC.  I *assume* that if we can get the
shells playing nice, these apps will fall in line, as they're launched by a

At any rate, as I said before, my understanding is limited and I
would appreciate some enlightenment from those with greater comprehension.

As a final data point, I'm testing these changes by adding the requisite
lines to my ~/.Xresources, doing a xrdb -merge ~/.Xresources, and launcing
a fresh xterm for testing.  If this procedure is incorrect, as section 3.8
of policy may be obliquely indicating, someone please inform me of the
Right Way.

I'm not going to push anything into XFree86 until I can be reasonably sure
it's going to work, and I've tested it myself.

G. Branden Robinson                 |
Purdue University                   |           //     // //  /     /
branden@purdue.edu                  |           EI 'AANIIGOO 'AHOOT'E
http://www.ecn.purdue.edu/~branden/ |

Attachment: pgpetlX_mg0rZ.pgp
Description: PGP signature

Reply to: