Thread (10 messages) 10 messages, 4 authors, 2011-03-25

Re: [PATCH v3] Documentation: Add evdev type and code definitions

flat view

From: Peter Hutterer <hidden>
Date: 2011-03-25 06:04:52
Also in: lkml

On Thu, Mar 24, 2011 at 10:50:23PM -0700, Dmitry Torokhov wrote:
On Fri, Mar 25, 2011 at 10:36:02AM +1000, Peter Hutterer wrote:
quoted
On Thu, Mar 24, 2011 at 03:45:51PM -0400, Chase Douglas wrote:
quoted
On 03/24/2011 02:44 PM, Chris Bagwell wrote:
quoted
Looking good.  Minor comments at end.

On Thu, Mar 24, 2011 at 12:44 PM, Chase Douglas
[off-list ref] wrote:
quoted
+Touchscreens:
+----------
+ABS_{X,Y} must be reported with the location of the touch. BTN_TOUCH must be
+used to report when a touch is active on the screen.
+BTN_{MOUSE,LEFT,MIDDLE,RIGHT} must not be reported. BTN_TOOL_<name> events
+should be reported where possible.
Does the real button event restriction solve something?  I know
majority of touchscreens are not sending a button today but usually
there is a button on the touchscreen frame that I'd image at least
some devices hooking it up to touchscreen interface.

stylus+touchscreen combo devices are surely going to want to do this.
Perhaps this should be more clearly for touchscreen-only devices (not
tablets or a mix). The problem we end up seeing is drivers who use
BTN_LEFT because they want a touch to generate a primary button press in
X. I want to prevent people from doing this.
suggest that they use BTN_0 instead then?
Umm, I'd hope X evdev driver could handle BTN_TOUCH without
BTN_LEFT/BTN_0 to indicate primary button. Legacy mode (mousedev)
already translateds BTN_TOUCH/BTN_LEFT/BTN_0 into "left PS/2 button".
evdev treats BTN_TOUCH as button, yes. my comment above refers to additional
buttons on the tablet that aren't touch. maybe I misread the above though.

Cheers,
  Peter
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help