Closed lojjic closed 4 years ago
No worries! As conflicts go this is an easy one to unwrap, but thank you for the resolution tips.
Your other comments all make perfect sense, I'll work through them shortly.
Conflicts are resolved (via rebase) and all the fixes from that first round of code review have been made.
I've now tested this with the webxr-samples
repo without issue, but still welcome additional testcase sources. As well as hints around how to test the session suspension logic.
Looks great! Thank you so much!
Fixes:
xrSession.requestAnimationFrame
to use a single device frame handler and a list of callbacks, to match the spec algorithm and prevent multiple vrDisplay.submitFrame calls per frameactive
andanimationFrame
booleans to XRFrame, toggle them appropriately, and check them during pose population per specNotes for Review:
test-xr-session.js
test file is busted so I haven't added anything there to test these changes, but I'd feel much better if I could, so I may try to at least get that one into a runnable state so I can add to it.suspendedCallback
in favor of the callbacks array, and I think the logic around suspension (pausing/restarting the device frame loop) is correct, but I don't know how to actually test the suspension. Tips welcome.