Re: AltGr finger twisters documented? - bug in keyboard handling?

[Juanma Barranquero <lekktu@gmail.com>]

On 7/5/05, Lennart Borgman <lennart.borgman.073@student.lu.se> wrote:
> Juanma Barranquero wrote:
> 
> >Exactly. With `w32-recognize-altgr' set to nil there's no way I can
> >type C-\, because there's no way I can type \ directly.

As I've just said in another message, I was wrong on this account. I
can type \ as LCtrl-AltGr-º

>     w32-recognize-altgr's value is t
> 
>     Recognize right-alt and left-ctrl as AltGr.
>     When nil, the right-alt and left-ctrl key combination is
>     interpreted normally.

> I can not see that left-ctrl works as AltGr. That is good, I would not
> want it to, but the doc string should be changed I believe.

I'm not sure the documentation is wrong, though it is less than clear.

My interpretation (but I can be very wrong, as I've not looked at the
code) is that the Windows keyboard code translates AltGr to keyboard
events LCtrl + RAlt. Emacs intercepts this behavior (at least with
w32-pass-alt-to-system set to nil, which is almost necessary to use
Emacs normally).  When w32-recognize-altgr is t, Emacs is emulating
the Windows AltGr behaviour, and that's why AltGr and LCtrl-AltGr do
the same.

> The other, more serious thing, is that different keyboards seems to
> behave differently. They are not supposed to do that I guess. Or am I
> wrong for some reason?

Really, all data seen in this thread seem to point that:

 - It is Windows specific
 - Does not depend on the keyboard
  - *Does* depend on `w32-recognize-altgr'

> Maybe all this just relates to w32? Anyway it would be very good if it
> was fixed before release!

I really don't think there's any bug to fix. Only to document.

-- 
                    /L/e/k/t/u