Screwed up keyboard mappings on RedHat

Screwed up keyboard mappings on RedHat

[Roy Smith]

I'm using GNU Emacs 21.2.1 on a RedHat 8.0 box.  Neither my delete nor 
C-H key are working right.  C-H is acting like the delete key should.  
When I type "M-X help return c C-H", I get:

DEL runs the command backward-delete-char-untabify

Obviously something is mapping the C-H I typed into DEL.  I know it's 
not happening in some terminal emulation or transport layer before emacs 
gets hold of it.  If I trace the emacs process, I see it reading the 
right characters from stdin:

[roy@bonkers roy]$ grep '^read(0' typescript 
read(0, "\10", 1)                       = 1
read(0, "\10", 1)                       = 1
read(0, "\33x", 2)                      = 2
read(0, "h", 1)                         = 1
read(0, "e", 1)                         = 1
read(0, "l", 1)                         = 1
read(0, "p", 1)                         = 1
read(0, "\r", 1)                        = 1
read(0, "c", 1)                         = 1
read(0, "\10", 1)                       = 1
read(0, "\30", 1)                       = 1
read(0, "\3", 1)                        = 1

Two questions:

1) Where is this bogus mapping being done?

2) Why did RedHat screw this up?

Re: Screwed up keyboard mappings on RedHat

[Kai Großjohann]

Roy Smith <roy@panix.com> writes:

> I'm using GNU Emacs 21.2.1 on a RedHat 8.0 box.  Neither my delete nor 
> C-H key are working right.  C-H is acting like the delete key should.  
> When I type "M-X help return c C-H", I get:
>
> DEL runs the command backward-delete-char-untabify

Interesting.  Investigate the variable normal-erase-is-backspace and
the function normal-erase-is-backspace-mode.  Does this help?

-- 
file-error; Data: (Opening input file no such file or directory ~/.signature)

Re: Screwed up keyboard mappings on RedHat

[Roy Smith]

kai.grossjohann@gmx.net (Kai Grossjohann) wrote:

> Roy Smith <roy@panix.com> writes:
> 
> > I'm using GNU Emacs 21.2.1 on a RedHat 8.0 box.  Neither my delete nor 
> > C-H key are working right.  C-H is acting like the delete key should.  
> > When I type "M-X help return c C-H", I get:
> >
> > DEL runs the command backward-delete-char-untabify
> 
> Interesting.  Investigate the variable normal-erase-is-backspace and
> the function normal-erase-is-backspace-mode.  Does this help?

Hmmm.  It's currently set to "t".  It looks like that might be what I 
need to play with, thanks.

In the meantime (before your response came), I had already discovered 
keyboard translation tables and fixed the problem with:

(keyboard-translate ?\C-h nil)
(keyboard-translate ?\C-? nil)

in my .emacs file.  It gets me the behavior I want, even if it's not the 
right way to do it.

PS -- I've been using emacs for 20 years (since the original TOPS-10 
TECO version) and never even knew keyboard translation tables existed.  
Is this a new development?

Re: Screwed up keyboard mappings on RedHat

[Kevin Rodgers]

Roy Smith wrote:

> PS -- I've been using emacs for 20 years (since the original TOPS-10 
> TECO version) and never even knew keyboard translation tables existed.  
> Is this a new development?

Nope.  Here are the oldest references I can find in the ChangeLog* files:


Mon Jan  2 02:49:22 1989  Richard Stallman  (rms@mole.ai.mit.edu)
...
	* vt200.el (keyboard-translate-table): Make a table,
	and use it to interchange ESC and backquote.

Sat Mar 10 17:19:38 1990  Joseph Arceneaux  (jla at churchy.ai.mit.edu)
...
	* keyboard.c (get_char):  Re-set obj if we've keyboard-translated
	c.

Fri Oct 26 00:13:25 1990  Richard Stallman  (rms@mole.ai.mit.edu)
...
	* subr.el (keyboard-translate): New function.

-- 
<a href="mailto:&lt;kevin.rodgers&#64;ihs.com&gt;">Kevin Rodgers</a>

Re: Screwed up keyboard mappings on RedHat

[Kevin Rodgers]

Kevin Rodgers wrote:

> Roy Smith wrote:
>> PS -- I've been using emacs for 20 years (since the original TOPS-10 
>> TECO version) and never even knew keyboard translation tables 
>> existed.  Is this a new development?
> 
> Nope.  Here are the oldest references I can find in the ChangeLog* files:
> 
> 
> Mon Jan  2 02:49:22 1989  Richard Stallman  (rms@mole.ai.mit.edu)
> ...
>     * vt200.el (keyboard-translate-table): Make a table,
>     and use it to interchange ESC and backquote.
> 
> Sat Mar 10 17:19:38 1990  Joseph Arceneaux  (jla at churchy.ai.mit.edu)
> ...
>     * keyboard.c (get_char):  Re-set obj if we've keyboard-translated
>     c.
> 
> Fri Oct 26 00:13:25 1990  Richard Stallman  (rms@mole.ai.mit.edu)
> ...
>     * subr.el (keyboard-translate): New function.

And since you're a real old-timer, here's a reference from the *NEWS files (I

can only trace my Emacs experience back to a TOPS-20 implementation -- not
sure if it was TECO or not :-)


Changes in Emacs 1.12
...
* You can now specify a translation table for keyboard
  input characters, as a way of exchanging or substituting
  keys on the keyboard.

-- 
<a href="mailto:&lt;kevin.rodgers&#64;ihs.com&gt;">Kevin Rodgers</a>

Re: Screwed up keyboard mappings on RedHat

[Kai Großjohann]

Roy Smith <roy@panix.com> writes:

> Hmmm.  It's currently set to "t".  It looks like that might be what I 
> need to play with, thanks.
>
> In the meantime (before your response came), I had already discovered 
> keyboard translation tables and fixed the problem with:
>
> (keyboard-translate ?\C-h nil)
> (keyboard-translate ?\C-? nil)

Right.  That's basically the effect of
(normal-erase-is-backspace-mode -1) :-)
-- 
file-error; Data: (Opening input file no such file or directory ~/.signature)