On Sun, Sep 23, 2001 at 12:45:42AM +0200, Antonio Larrosa Jiménez wrote:
> First of all, please understand that this request was done independently
> of the patch submitted. The patch still has to be applied even if Button6
> and 7 are not added to X.h .
I think forking the core protocol is frowned upon.
> I suppose that means I'll have to implement a XInput extension to be used
> with usual mice and trackballs. Note that I've never done anything like
> that before (just another patch some time ago for mouse.c), so I'd like to
> know if there's any documentation on how to do so, or I should just read
> similar things like the wacom driver and just duplicate most of the mouse
> driver.
That's probably a good place to start.
> Also, is it really possible to have a normal mouse driver for a mouse and
> then using an XInput driver for the same device in order to be able to
> send more button events ? I'd say this is needed, so that "old"
> applications can receive normal mouse events while newer applications will
> be able to understand the new XInput driver with the horizontal wheel,
> isn't it so? Is it really possible to have two drivers to handle the same
> device?
Any XInput device can be configured as the "core pointer", so I wouldn't
worry about this.
> >From another point of view, why are there two options since a long time
> ago for the mouse driver called XAxisMapping and YAxisMapping each one
> taking two button numbers ? If we have 3 standard button numbers, aren't
> these options supposed to be used for buttons number 4,5,6 and 7? If 6 and
> 7 shouldn't be used, why are those options there at first place taking 4
> different numbers?
Don't know that one. Maybe for Macintosh mice with two wheels? :)
--
G. Branden Robinson | What influenced me to atheism was
Debian GNU/Linux | reading the Bible cover to cover.
[EMAIL PROTECTED] | Twice.
http://people.debian.org/~branden/ | -- J. Michael Straczynski
PGP signature
