On Tue, Feb 28, 2012 at 09:01:44AM +0000, Przemo Firszt wrote: > Dnia 2012-02-27, pon o godzinie 09:55 -0800, Jason Gerecke pisze: > [..] > > > > "Dead" keys are a result of both the block of code you mentioned and > > the sendAButton function using the *mapped* button number (instead of > > the *physical* button number) as an index into the actions array. When > > xsetwacom sets "button 4", its setting the action to be performed when > > X11 button 4 is used. Since the block of code above maps no physical > > button to X11 button 4, it winds up "dead". Actually, if you can find > > a way of changing the priv->buttons array (they exist, but aren't in > > common use thankfully) you can make buttons perform the actions you > > associated with another button... > > > > Personally I'd like to see this behavior go away. Skipping buttons 4-7 > > is one thing (since it doesn't really make sense to have the tablet > > buttons produce scroll events by default), but having them be "dead" > > just leads to confusion. The only thing that makes me hesitant is the > > fact that it'd break *every* configuration script that sets a button > > higher than 3. It'd be annoying for users to fix, but probably not a > > huge issue. > > > > Thoughts? > > > I think we might wait for Peter and Ping. I'd like to see it go away as > well, but hid drive is still far from finished and the behaviour is > consistend with usb driver. I might come back to it when hid-wacom is > finished.
I assume the reason you are confused is because you are working too much with the kernel and its BTN_4 numbering ;) realistically, we cannot get rid of it. button 4-7 for scrolling will stay with us as long as X is alive. I think the issue here is more one of documentation than an actual technical issue. even in the default configuration, a user never sees a button 4 press event (except for scrolling). so to the user, button 4 is already button 8. The best argument I can find is that we're using the logical button mapping in the driver as opposed to e.g. evdev's physical button mapping in the driver and that is inconsistent between drivers. Not the only thing inconsistent though. other than that, right now I don't see what benefit it really gives us, especially for the cost of breaking scripts. Plus, if Button 4-7 is dead, it should be easy enough to put out an xsetwacom warning and refer users to the man page. Cheers, Peter ------------------------------------------------------------------------------ Virtualization & Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ _______________________________________________ Linuxwacom-devel mailing list Linuxwacom-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxwacom-devel
