Relative pointer protocol

[ids1024]

I wonder if any of this could be simpler and less redundant...

Needs testing and probably some corrections.

[ids1024]

I'm not sure how best to test this protocol on it's own. I think in first person games it tends to be used together with zwp_pointer_constraints_v1. Need to see about implementing that one.

[Drakulix]

I'm not sure how best to test this protocol on it's own.

Maybe the wine-wayland-patchset supports it?

Also wlcs actually seems to have some relative-pointer tests, if you feel like enabling those tests for wlcs_anvil.
https://github.com/MirServer/wlcs/blob/main/src/relative_pointer_unstable_v1.cpp

[ids1024]

Enabled the wlcs test for this, though it seems fairly rudimentary so it isn't really a complete test for relative pointer functionality.

[codecov-commenter]

Codecov Report

Base: 25.27% // Head: 25.34% // Increases project coverage by +0.07% 🎉

Coverage data is based on head (3cfdcd7) compared to base (ffbc84e).
Patch coverage: 34.13% of modified lines in pull request are covered.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #757      +/-   ##
==========================================
+ Coverage   25.27%   25.34%   +0.07%     
==========================================
  Files         130      131       +1     
  Lines       20882    21123     +241     
==========================================
+ Hits         5277     5354      +77     
- Misses      15605    15769     +164     

Flag	Coverage Δ
wlcs-core	22.41% <2.81%> (-0.24%)	⬇️
wlcs-output	10.28% <2.81%> (-0.10%)	⬇️
wlcs-pointer-input	24.60% <34.13%> (+0.08%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
anvil/src/input_handler.rs	9.02% <0.00%> (ø)
anvil/src/shell/grabs.rs	1.54% <0.00%> (-0.16%)	⬇️
src/backend/input/mod.rs	0.00% <0.00%> (ø)
src/desktop/wayland/layer.rs	9.49% <0.00%> (-0.08%)	⬇️
src/desktop/wayland/popup/grab.rs	0.00% <0.00%> (ø)
src/input/mod.rs	52.63% <ø> (ø)
src/wayland/data_device/dnd_grab.rs	0.00% <0.00%> (ø)
src/wayland/data_device/server_dnd_grab.rs	0.00% <0.00%> (ø)
src/wayland/relative_pointer.rs	0.00% <0.00%> (ø)
src/wayland/seat/pointer.rs	19.17% <30.76%> (+1.56%)	⬆️
... and 10 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[ids1024]

I've removed the somewhat incomplete pointer constraints implementation, and the remaining issues here should be fixed now. Pointer constraints can be added in a later PR.

Real use cases seem to typically use this with pointer constraints, but https://github.com/Smithay/client-toolkit/blob/master/examples/relative_pointer.rs can demonstrate that a client receives the correct events.

[PolyMeilex]

One nitpick, but in general looks reasonable to me.

[Drakulix]

Looks good to me, expect utime vs time. Can we unify those? What makes the relative-motion event have a higher precision?