NTEmacs and C-M -x cords

NTEmacs and C-M -x cords

[Le Wang]

[-- Attachment #1: Type: text/plain, Size: 307 bytes --]

Does pressing C-hk, <left control>+<right alt>+x work for anyone else?  All
left control and right alt cords are broken for me;  left shift and right
alt is fine.  I've tried official build of 23.2.1 with Vista and XP.  I've
also tried CVS Emacs 24 without success.  Does this work for anyone else?

-- 
Le

[-- Attachment #2: Type: text/html, Size: 355 bytes --]

Re: NTEmacs and C-M -x cords

[Le Wang]

[-- Attachment #1: Type: text/plain, Size: 552 bytes --]

On Wed, Feb 16, 2011 at 8:02 PM, Le Wang <l26wang@gmail.com> wrote:

> Does pressing C-hk, <left control>+<right alt>+x work for anyone else?  All
> left control and right alt cords are broken for me;  left shift and right
> alt is fine.  I've tried official build of 23.2.1 with Vista and XP.  I've
> also tried CVS Emacs 24 without success.  Does this work for anyone else?
>

An additional note, I know that the keystroke makes it through to programs
(i.e. nothing is capturing keys), because I can bind  this cord to a
function in MSWord.


-- 
Le

[-- Attachment #2: Type: text/html, Size: 838 bytes --]

Re: NTEmacs and C-M -x cords

[B. T. Raven]

Le Wang wrote:
> On Wed, Feb 16, 2011 at 8:02 PM, Le Wang <l26wang@gmail.com> wrote:
> 
>> Does pressing C-hk, <left control>+<right alt>+x work for anyone else?  All
>> left control and right alt cords are broken for me;  left shift and right
>> alt is fine.  I've tried official build of 23.2.1 with Vista and XP.  I've
>> also tried CVS Emacs 24 without success.  Does this work for anyone else?
>>
> 
> An additional note, I know that the keystroke makes it through to programs
> (i.e. nothing is capturing keys), because I can bind  this cord to a
> function in MSWord.
> 
> 

Are you sure they're supposed to work? What's wrong with using left alt
with left control? It's a stretch (literally) to call those combos
chords. Here none of them produces anything except with key "j" which
for some reason raises an instance of Nero Showtime (same as with left
alt-right control) other combos with LAlt-RCtl seem to be seen by Emacs.
I am using Keytweak but that shouldn't affect anything in Emacs since it
remaps scancodes in the registry.

Ed

Re: NTEmacs and C-M -x cords

[Le Wang]

[-- Attachment #1: Type: text/plain, Size: 381 bytes --]

On Wed, Feb 16, 2011 at 10:12 PM, B. T. Raven <nihil@nihilo.net> wrote:

> Are you sure they're supposed to work?
>


 I'm not sure.  I haven't seen anything that leads me to believe
LControl+RAlt being ignored is by design.  It certainly doesn't seem right
to me.

What's wrong with using left alt with left control?
>


I want to use RAlt more for ergonomic reasons.


-- 
Le

>

[-- Attachment #2: Type: text/html, Size: 1213 bytes --]

RE: NTEmacs and C-M -x cords

[Drew Adams]

>> Are you sure they're supposed to work?
>
> I'm not sure.  I haven't seen anything that leads me to believe
> LControl+RAlt being ignored is by design.  It certainly doesn't
> seem right to me. I want to use RAlt more for ergonomic reasons.

It might or might not be possible for Emacs to be fixed in this regard.
Consider filing a bug report.  Emacs developers will clarify whether they
consider this a bug.    Perhaps such a limitation should be documented, to save
wondering by users - dunno.

FWIW, I see the same thing on Windows, in Emacs 20 through 24.  Emacs doesn't
seem to receive the key combination at all - `open-dribble-file' doesn't record
it.  There might be nothing that Emacs can do about this.  Dunno.

Re: NTEmacs and C-M -x cords

[Le Wang]

[-- Attachment #1: Type: text/plain, Size: 69 bytes --]

Filed bug: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=8058

-- 
Le

[-- Attachment #2: Type: text/html, Size: 174 bytes --]

Re: NTEmacs and C-M -x cords

[Deniz Dogan]

2011/2/16 Le Wang <l26wang@gmail.com>:
> Does pressing C-hk, <left control>+<right alt>+x work for anyone else?  All
> left control and right alt cords are broken for me;  left shift and right
> alt is fine.  I've tried official build of 23.2.1 with Vista and XP.  I've
> also tried CVS Emacs 24 without success.  Does this work for anyone else?
>

Left control + AltGr + x doesn't do anything for me in Emacs 24. This
is with a Swedish keyboard, where AltGr is located where the right Alt
usually is.

-- 
Deniz Dogan

Re: NTEmacs and C-M -x cords

[Eli Zaretskii]

> Date: Wed, 16 Feb 2011 20:02:16 +0800
> From: Le Wang <l26wang@gmail.com>
> 
> Does pressing C-hk, <left control>+<right alt>+x work for anyone else?  All
> left control and right alt cords are broken for me;  left shift and right
> alt is fine.  I've tried official build of 23.2.1 with Vista and XP.  I've
> also tried CVS Emacs 24 without success.  Does this work for anyone else?

For the record: this is by design, see the variable
w32-recognize-altgr, which allows to customize this behavior.

RE: NTEmacs and C-M -x cords

[Drew Adams]

> For the record: this is by design, see the variable
> w32-recognize-altgr, which allows to customize this behavior.

Good to know.

How about indexing Emacs-manual node `Windows Keyboard' (which covers this)
using some entries that include `alt', `ctrl', and `control'.  It is likely that
some users will search for info about those keys.

`w32-recognize-altgr' itself is an index entry to that node, but someone who
doesn't have an AltGr key might not know about it and might wonder why left ctrl
+ right alt doesn't seem to do anything (by default).

Also, `w32-recognize-altgr' is not a user option, so your "customize this
behavior" does not include using Customize.  How about making it a user option?

Re: NTEmacs and C-M -x cords

[Eli Zaretskii]

> From: "Drew Adams" <drew.adams@oracle.com>
> Cc: <help-gnu-emacs@gnu.org>
> Date: Wed, 16 Feb 2011 10:45:50 -0800
> 
> How about indexing Emacs-manual node `Windows Keyboard' (which covers this)
> using some entries that include `alt', `ctrl', and `control'.  It is likely that
> some users will search for info about those keys.

But "i alt RET" will land you there anyway, so adding another "alt"
entry is not useful, or needed.

> `w32-recognize-altgr' itself is an index entry to that node, but someone who
> doesn't have an AltGr key might not know about it and might wonder why left ctrl
> + right alt doesn't seem to do anything (by default).

But "i alt RET" will land you there anyway.

> Also, `w32-recognize-altgr' is not a user option, so your "customize this
> behavior" does not include using Customize.  How about making it a user option?

Patches for that are welcome.

RE: NTEmacs and C-M -x cords

[Drew Adams]

> > How about indexing Emacs-manual node `Windows Keyboard' 
> > (which covers this) using some entries that include `alt',
> > `ctrl', and `control'.  It is likely that
> > some users will search for info about those keys.
> 
> But "i alt RET" will land you there anyway, so adding another "alt"
> entry is not useful, or needed.

Somehow I expected that response from you. ;-)

The fact that "alt RET" works does not mean that an entry for the right alt key
would not _also_ be helpful.  E.g., "right alt key".

And as I suggested, entries for "ctrl" and "control" might also help here.

Think of Le's question as an index search: "Why doesn't left-control plus
right-alt work for `C-M-'?".

That's a reasonable question, especially given that using left keys for both
control and alt and using right keys for both of them work as expected - each
combination is interpreted as `C-M-'.  And just the right alt key alone also
works, as `M-'.  It is only the combination left control plus right alt that
Emacs treats by default as control + altgr.

> > Also, `w32-recognize-altgr' is not a user option, so your 
> > "customize this behavior" does not include using Customize.
> > How about making it a user option?
> 
> Patches for that are welcome.

(defcustom w32-recognize-altgr t
  "Non-nil means treat keys right-alt and left-ctrl used together as AltGr.
If nil, treat the combination the same as ctrl plus alt (typically
`C-M-')."
  :group 'keyboard :type 'boolean)

For any C code (the defvar is in `w32term.c'), someone else will need to
contribute.