https://bugs.freedesktop.org/show_bug.cgi?id=106036
--- Comment #4 from peter.ganzh...@gmail.com ---
Thanks for the details, I think I understood how it should/can be done.
I added the following:
90-libinput-model-quirks.hwdb:
libinput:mouse:b0003v04B3p3100*
libinput:mouse:b0003v04B3p3103*
libinput:mouse:b0003v04B3p3105*
libinput:mouse:b0003v04B3p3108*
libinput:mouse:b0003v04B3p3109*
libinput:mouse:b0003v17EFp6049*
LIBINPUT_MODEL_SCROLLPOINT=1
evdev.h:
added EVDEV_MODEL_SCROLLPOINT = (1 << 31), to enum evdec_device_model
evdec.c:
added MODEL(SCROLLPOINT), to model_map in evdev_read_model_flags()
which should make me able to add a quirk to fallback_process_relative().
For quick testing I just added
if (device->model_flags & EVDEV_MODEL_SCROLLPOINT) { /* PGZH */
if (e->code == REL_WHEEL || e->code == REL_HWHEEL ) {
if ( e->value < 16 ) e->value=1;
if ( e->value < 32 && e->value >= 16 ) e->value=2;
if ( e->value < 63 && e->value >= 32 ) e->value=4;
if ( e->value == 63 ) e->value=8;
}
}
before e->code is evaluated and e->value is added to the dispatch->wheel.xy
values.
I assumed the value range to be in between 0 and 63 since this is what I see in
evemu-record.
Is this assumption correct? With the outlined changes and replacment of my
libinput.so (.10.13.0 and symlinks to .so.10 and .so of course) after building
and putting the 90-libinput-model-quirks.hwdb into /lib/udev/hwdb I don't
notice any change. Also setting e->value=1 in any case does not change
anything, so something must still be wrong.
Is the assumed range of e->value wrong (and how can I determine the correct
range easily if required) or what could still be wrong here?
--
You are receiving this mail because:
You are the assignee for the bug._______________________________________________
wayland-bugs mailing list
wayland-bugs@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/wayland-bugs
