23.0.50; delete vs deletechar

23.0.50; delete vs deletechar

[raman]

I suspect the multi-tty patch again, but [delete] has now changed
to deletechar.

I dont have a problem with this so long as we document that as
the truth --I'll happily update keybinding definitions in
emacspeak. Ditto with [insert] vs [insertchar]


In GNU Emacs 23.0.50.3 (i686-pc-linux-gnu, GTK+ Version 2.8.20, multi-tty)
 of 2007-08-30 on labrador
configured using `configure  '--prefix=/usr/local' '--with-gtk' '--with-gpm' '--with-rsvg' '--with-png' '--with-jpeg' '--with-gif''

Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: nil
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: nil
  value of $LANG: en_US.UTF-8
  locale-coding-system: utf-8
  default-enable-multibyte-characters: nil

Major mode: VM

Minor modes in effect:
  t-mouse-mode: t
  erc-track-minor-mode: t
  jabber-activity-mode: t
  dynamic-completion-mode: t
  dired-omit-mode: t
  shell-dirtrack-mode: t
  savehist-mode: t
  display-time-mode: t
  desktop-save-mode: t
  auto-insert-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  unify-8859-on-encoding-mode: t
  utf-translate-cjk-mode: t
  auto-compression-mode: t

Recent input:
f i x SPC i t SPC f o r SPC m y s e l f . RET RET H 
o w e v e r , SPC s p e a k i n g SPC f r o m SPC t 
h e SPC p o i n t SPC o f SPC ESC DEL ESC DEL v i e 
w p o i n t SPC o f SPC E m a c s p e a k SPC u s e 
r s , SPC m a n y SPC o f SPC w h o m SPC t o SPC ESC 
DEL SPC c o m e SPC t o SPC E m a c s SPC p u r e l 
y SPC b e c a u s e SPC E m a c s p e a k SPC i s SPC 
t h e i r SPC L i n u x SPC a c c e s s SPC s o l u 
t i o n , SPC i t SPC w o u l d SPC b e SPC n i c e 
SPC t o SPC h a v e SPC r e p l a c e - s t r i n g 
SPC b o u n d SPC t o SPC s o m e SPC k e y SPC - - 
- SPC I SPC c a r e SPC n o t SPC w h a t . SPC C-c 
C-c d n k k k k k k k k k d n k k k k j ' k k k k ESC 
x r e p o r t - e m TAB RET b TAB RET

Recent messages:
Decoding quoted-printable... done
Wrote /tmp/vm-pgg-signature26308evS
Decoding MIME message... done
Hid 7  citation lines
Decoding MIME message...
Decoding quoted-printable... done
Decoding MIME message... done
Decoding MIME message... done
Making completion list...
Loading emacsbug...done

-- 
Best Regards,
--raman

      
Email:  raman@users.sf.net
WWW:    http://emacspeak.sf.net/raman/
AIM:    emacspeak       GTalk: tv.raman.tv@gmail.com
PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
Google: tv+raman 
IRC:    irc://irc.freenode.net/#emacs

Re: 23.0.50; delete vs deletechar

[Glenn Morris]

raman <raman@users.sf.net>, "T. V. Raman" wrote:

> I suspect the multi-tty patch again, but [delete] has now changed
> to deletechar.

Not sure what you mean. In Emacs 22 I get:

C-d (translated from <delete>) runs the command delete-char
  which is an interactive built-in function in `C source code'.
It is bound to C-d, <deletechar>.
(delete-char N &optional KILLFLAG)

In Emacs CVS before and after multi-tty merge, I get the same, except
that C-d, <deletechar> have changed order. Does this matter?

C-d (translated from <delete>) runs the command delete-char
  which is an interactive built-in function in `C source code'.
It is bound to <deletechar>, C-d.
(delete-char N &optional KILLFLAG)

Re: 23.0.50; delete vs deletechar

[T. V. Raman]

On the Linux console:

Press the <delete> key (the one under the insert key)
running on the console.

In emacs22 that produces [delete] 
in emacs 23 that produces [<deletechar>]

Similarly for [insert] vs [<insertchar>]

Does this matter?

Well, of course it does if your code binds these keys in other
keymaps e.g. emacspeak uses C-e <delete> and c-e<insert> as
special keys; now I need to rebind them appropriately.

>>>>> "GM" == Glenn Morris <rgm@gnu.org> writes:
    GM> raman <raman@users.sf.net>, "T. V. Raman" wrote:
    >> I suspect the multi-tty patch again, but [delete] has now
    >> changed to deletechar.
    GM> 
    GM> Not sure what you mean. In Emacs 22 I get:
    GM> 
    GM> C-d (translated from <delete>) runs the command
    GM> delete-char which is an interactive built-in function in
    GM> `C source code'.  It is bound to C-d, <deletechar>.
    GM> (delete-char N &optional KILLFLAG)
    GM> 
    GM> In Emacs CVS before and after multi-tty merge, I get the
    GM> same, except that C-d, <deletechar> have changed
    GM> order. Does this matter?
    GM> 
    GM> C-d (translated from <delete>) runs the command
    GM> delete-char which is an interactive built-in function in
    GM> `C source code'.  It is bound to <deletechar>, C-d.
    GM> (delete-char N &optional KILLFLAG)

-- 
Best Regards,
--raman

      
Email:  raman@users.sf.net
WWW:    http://emacspeak.sf.net/raman/
AIM:    emacspeak       GTalk: tv.raman.tv@gmail.com
PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
Google: tv+raman 
IRC:    irc://irc.freenode.net/#emacs

Re: 23.0.50; delete vs deletechar

[Glenn Morris]

"T. V. Raman" wrote:

> On the Linux console:

Oh, of course.

> Press the <delete> key (the one under the insert key)
> running on the console.
>
> In emacs22 that produces [delete] 
> in emacs 23 that produces [<deletechar>]

For me, it produces deletechar in 22 as well.

Re: 23.0.50; delete vs deletechar

[T. V. Raman]

interesting, on 22. it always produced [delete] and [insert] for me.

>>>>> "GM" == Glenn Morris <rgm@gnu.org> writes:
    GM> "T. V. Raman" wrote:
    >> On the Linux console:
    GM> 
    GM> Oh, of course.
    GM> 
    >> Press the <delete> key (the one under the insert key)
    >> running on the console.
    >> 
    >> In emacs22 that produces [delete] in emacs 23 that
    >> produces [<deletechar>]
    GM> 
    GM> For me, it produces deletechar in 22 as well.

-- 
Best Regards,
--raman

      
Email:  raman@users.sf.net
WWW:    http://emacspeak.sf.net/raman/
AIM:    emacspeak       GTalk: tv.raman.tv@gmail.com
PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
Google: tv+raman 
IRC:    irc://irc.freenode.net/#emacs

Re: 23.0.50; delete vs deletechar

[Richard Stallman]

After reading the recent messages, I am not sure whether people agree
there is indeed now a problem in the decision about which character
should delete backwards.

But it occurs to me that when using multiple terminals, it could
happen that some of them want BACKSPACE to delete backwards and some
want DELETE to do it.

So shouldn't normal-erase-is-backspace-mode
operate on local-function-key-map?

Re: 23.0.50; delete vs deletechar

[T. V. Raman]

To clarify, I spotted two issues that are tangled up -- here is
an attempt to separate them.

  1. What should pressing the <delete> key on the linux console
    deliver? [delete] or [deletechar] I ask this independent of
    the command that that key invokes --- purely for knowing what
    should be used in lisp code that binds that key.
  2. When one launches Emacs on the console,
     and then runs emacsclient at a different console,
     should the new frame inherit normal-backspace-is-erase ? I
    believe it should -- it currently does not.
    
  

>>>>> "Richard" == Richard Stallman <rms@gnu.org> writes:
    Richard> After reading the recent messages, I am not sure
    Richard> whether people agree there is indeed now a problem
    Richard> in the decision about which character should delete
    Richard> backwards.
    Richard> 
    Richard> But it occurs to me that when using multiple
    Richard> terminals, it could happen that some of them want
    Richard> BACKSPACE to delete backwards and some want DELETE
    Richard> to do it.
    Richard> 
    Richard> So shouldn't normal-erase-is-backspace-mode operate
    Richard> on local-function-key-map?

-- 
Best Regards,
--raman

      
Email:  raman@users.sf.net
WWW:    http://emacspeak.sf.net/raman/
AIM:    emacspeak       GTalk: tv.raman.tv@gmail.com
PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
Google: tv+raman 
IRC:    irc://irc.freenode.net/#emacs

Re: 23.0.50; delete vs deletechar

[Richard Stallman]

      2. When one launches Emacs on the console,
	 and then runs emacsclient at a different console,
	 should the new frame inherit normal-backspace-is-erase ? 

I am not sure what that question means.  When you say "inherit", what
do you mean precisely?  Inherit from what?  Under what circumstances?

As far as I can see from the source code, this decision is a global
decision, and it applies to existing and future frames.

Re: 23.0.50; delete vs deletechar

[Dan Nicolaescu]

Richard Stallman <rms@gnu.org> writes:

  > After reading the recent messages, I am not sure whether people agree
  > there is indeed now a problem in the decision about which character
  > should delete backwards.
  > 
  > But it occurs to me that when using multiple terminals, it could
  > happen that some of them want BACKSPACE to delete backwards and some
  > want DELETE to do it.
  > 
  > So shouldn't normal-erase-is-backspace-mode
  > operate on local-function-key-map?

Is there any evidence that normal-erase-is-backspace-mode is not
working properly?  
There is a new function normal-erase-is-backspace-setup-frame that is
called from make-frame, so it seems probable that it does setup things
correctly.

Re: 23.0.50; delete vs deletechar

[Richard Stallman]

    Is there any evidence that normal-erase-is-backspace-mode is not
    working properly?  

I don't know how to test it myself.  I just read the source code of
that function.

Reading it again, it seems to operate on local-function-key-map,
which seems right.

Re: 23.0.50; delete vs deletechar

[Stefan Monnier]

>     Is there any evidence that normal-erase-is-backspace-mode is not
>     working properly?  

> I don't know how to test it myself.  I just read the source code of
> that function.

> Reading it again, it seems to operate on local-function-key-map,
> which seems right.

That can only be right if normal-erase-is-backspace-mode is
a terminal-local variable.


        Stefan

Re: 23.0.50; delete vs deletechar

[Richard Stallman]

    That can only be right if normal-erase-is-backspace-mode is
    a terminal-local variable.

normal-erase-is-backspace-mode is a function, not a variable.
And it seems to be designed to affect the current terminal only.

normal-erase-is-backspace is a customizable variable, and I can't find
anything to make it terminal-local.  So I think customizing it
probably has a bug.

I do not know any way to test Emacs with multiple terminals.
Can someone else please test the function and the variable?
Then please reply with a report.

Re: 23.0.50; delete vs deletechar

[Richard Stallman]

[I sent this message a few weeks ago but did not get a response.
Could we get the discussion moving again?]

    That can only be right if normal-erase-is-backspace-mode is
    a terminal-local variable.

normal-erase-is-backspace-mode is a function, not a variable.
And it seems to be designed to affect the current terminal only.

normal-erase-is-backspace is a customizable variable, and I can't find
anything to make it terminal-local.  So I think customizing it
probably has a bug.

I do not know any way to test Emacs with multiple terminals.
Can someone else please test the function and the variable?
Then please reply with a report.

Re: 23.0.50; delete vs deletechar

[Dan Nicolaescu]

Richard Stallman <rms@gnu.org> writes:

  > [I sent this message a few weeks ago but did not get a response.
  > Could we get the discussion moving again?]
  > 
  >     That can only be right if normal-erase-is-backspace-mode is
  >     a terminal-local variable.
  > 
  > normal-erase-is-backspace-mode is a function, not a variable.
  > And it seems to be designed to affect the current terminal only.
  > 
  > normal-erase-is-backspace is a customizable variable, and I can't find
  > anything to make it terminal-local.  So I think customizing it
  > probably has a bug.
  > 
  > I do not know any way to test Emacs with multiple terminals.

You can do:

emacs -f server-start

and then:

emacsclient -t 

on one or more terminal.

  > Can someone else please test the function and the variable?
  > Then please reply with a report.


Disclaimer: I never understood all the details of
normal-erase-is-backspace (starting with the name).

I tested the function and it's effects seem to be correct, and it
applies only to the frame on the terminal where it is run from, it
does not affect the other terminal frames.

I don't know how to test the variable, I hope someone that knows how
it is supposed to work can do it.

Re: 23.0.50; delete vs deletechar

[Richard Stallman]

    I tested the function and it's effects seem to be correct, and it
    applies only to the frame on the terminal where it is run from, it
    does not affect the other terminal frames.

Maybe there is no bug.

Thanks for trying it.