Re: [E-devel] Multi-touch callbacks?

On Wed, 8 Sep 2021 14:40:22 -0500 David Grubb <[email protected]>
said:

no - that's fine! :) yes - if you have multiple screens and touch panels.. this
can be an issue as they may end up accidentally spanning all screens with the
single touch device.

> Hi Carsten,
> 
> Thanks for the prod, it helps to rule out any mistakes I'm making trying to
> test the receiving endpoint.
> 
> Turns out there wasn't an implementation issue on the application end. With
> both my test screens I needed to explicitly match them to libinput,
> otherwise they were treated as single point (mouse?) devices with some
> two-finger scrolling, e.g.,:
> 
> Section "InputClass"
>         Identifier "Planar"
>         MatchProduct "Weida"
>         Driver "libinput"
> EndSection
> 
> Sorry to bring platform config issues into the dev list. Appreciate the
> response. :)
> 
> Cheers,
> Dave.
> 
> 
> On Sun, Sep 5, 2021 at 8:11 AM Carsten Haitzler <[email protected]>
> wrote:
> 
> > On Sat, 4 Sep 2021 19:18:50 -0500 David Grubb <[email protected]
> > >
> > said:
> >
> > have you tried the elementary gesture layer tests?
> >
> > > I've been using EFL to build applications for a few years now and have
> > been
> > > very pleased with the results the libraries enable. I have a new
> > > requirement to implement pinch-and-zoom so I need to start exploring
> > > multitouch but I'm running into a few issues getting it working.
> > >
> > > I'm testing on my laptop which has a multitouch monitor and can confirm
> > > that the driver is emitting the correct kernel MT protocol events via
> > > evtest. For example, performing a small two finger drag I see two points
> > > reported and updated before release:
> > >
> > >
> > > Event: time 1630799195.877440, type 3 (EV_ABS), code 47 (ABS_MT_SLOT),
> > > value 0
> > > Event: time 1630799195.877440, type 3 (EV_ABS), code 57
> > > (ABS_MT_TRACKING_ID), value 2
> > > Event: time 1630799195.877440, type 3 (EV_ABS), code 53
> > > (ABS_MT_POSITION_X), value 6621
> > > Event: time 1630799195.877440, type 3 (EV_ABS), code 54
> > > (ABS_MT_POSITION_Y), value 3292
> > > Event: time 1630799195.877440, type 1 (EV_KEY), code 330 (BTN_TOUCH),
> > value
> > > 1
> > > Event: time 1630799195.877440, type 3 (EV_ABS), code 0 (ABS_X), value
> > 6621
> > > Event: time 1630799195.877440, type 3 (EV_ABS), code 1 (ABS_Y), value
> > 3292
> > > Event: time 1630799195.877440, -------------- SYN_REPORT ------------
> > > Event: time 1630799195.885821, type 3 (EV_ABS), code 47 (ABS_MT_SLOT),
> > > value 1
> > > Event: time 1630799195.885821, type 3 (EV_ABS), code 57
> > > (ABS_MT_TRACKING_ID), value 3
> > > Event: time 1630799195.885821, type 3 (EV_ABS), code 53
> > > (ABS_MT_POSITION_X), value 5302
> > > Event: time 1630799195.885821, type 3 (EV_ABS), code 54
> > > (ABS_MT_POSITION_Y), value 3490
> > > Event: time 1630799195.885821, -------------- SYN_REPORT ------------
> > > Event: time 1630799195.943159, type 3 (EV_ABS), code 57
> > > (ABS_MT_TRACKING_ID), value -1
> > > Event: time 1630799195.943159, -------------- SYN_REPORT ------------
> > > Event: time 1630799195.961918, type 3 (EV_ABS), code 47 (ABS_MT_SLOT),
> > > value 0
> > > Event: time 1630799195.961918, type 3 (EV_ABS), code 57
> > > (ABS_MT_TRACKING_ID), value -1
> > > Event: time 1630799195.961918, type 1 (EV_KEY), code 330 (BTN_TOUCH),
> > value
> > > 0
> > >
> > > Next, I compiled EFL with the flags: disable-tslib, enable-xinput22, and
> > > enable-always-build-examples. I made a few logging additions to the
> > >  evas_multi_touch example so it'd print on MULTI_x callbacks, but I'm
> > only
> > > seeing MOUSE_x callbacks. My understanding is that the first touch point
> > > would be reported as a MOUSE_DOWN/MOUSE_MOVE, and the second concurrent
> > > point would be reported as a MULTI_x. Is this correct, and am I missing
> > > something? I'm fairly confident that the driver itself is doing the
> > correct
> > > thing, but I'm a bit unsure about the interaction between EFL and X
> > (N.B.,
> > > the context is an OpenGL X window).
> > >
> > > If anybody has any suggestions or a direction they can point me from here
> > > I'd be very grateful.
> > >
> > > Cheers,
> > > Dave.
> > >
> > > _______________________________________________
> > > enlightenment-devel mailing list
> > > [email protected]
> > > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
> > >
> >
> >
> > --
> > ------------- Codito, ergo sum - "I code, therefore I am" --------------
> > Carsten Haitzler - [email protected]
> >
> >
> 
> _______________________________________________
> enlightenment-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
> 


-- 
------------- Codito, ergo sum - "I code, therefore I am" --------------
Carsten Haitzler - [email protected]