To track media devices, we were previously relying on a combination of LiveKit's useMediaDeviceSelect hook, and an object called UserChoices. Device settings should be accessible from outside a call, but the latter hook should only be used with a room or set of preview tracks, so it couldn't be raised to the app's top level. I also felt that the UserChoices code was hard to follow due to lack of clear ownership of the object.
To bring clarity to media device handling and allow device settings to be shown outside a call, I refactored these things into a single MediaDevicesContext which is instantiated at the top level of the app. Then, I had to manually sync LiveKit's device state with whatever is present in the context. This refactoring ended up fixing a couple other bugs with device handling along the way.
by fixing the cause rather than the symptom: this upgrades the code to use the new, recommended JSX transform mode of React 17+, which no longer requires you to import React manually just to write JSX.
* Change `jwt_service_url` to `livekit_service_url`
* Make it a POST so we can send the openID token sensibly
* Get an OIDC token & pass it with the request
* Read the SFU URL from there too
and convert the auth server accordingly, althugh with no actual OIDC
support yet, it just issues tokens blindly just as before and ignores
the openid token completely.
We'll need to update configs & the JWT service before merging this.
In preparation for adding layouts other than big grid to the NewVideoGrid component, I've abstracted the grid layout system into an interface called Layout. For now, the only implementation of this interface is BigGrid, but this will allow us to easily plug in Spotlight, SplitGrid, and OneOnOne layout systems so we can get rid of the old VideoGrid component and have One Grid to Rule Them All™.
Please do shout if any of this seems obtuse or underdocumented, because I'm not super happy with how approachable the NewVideoGrid code looks right now…
Incidentally, this refactoring made it way easier to save the state of the grid while in fullscreen / another layout, so I went ahead and did that.
We're now using LiveKit's magic RoomAudioRenderer component to make sure everyone's audio is rendered regardless of whether they have a tile in the DOM.
Calls are an environment with high cognitive load, so it's important that we keep extra UI elements like these to a minimum and stick to what's been explicitly designed. I assume that this was here as a developer feature to diagnose reliability of the back end components, which is perfectly fine, so I've kept it behind a developer setting rather than fully removing it.
* respect mute state set in lobby for call
Signed-off-by: Timo K <toger5@hotmail.de>
* move device from lobby to call
Signed-off-by: Timo K <toger5@hotmail.de>
* save device in local storage
Signed-off-by: Timo K <toger5@hotmail.de>
* local storage + fixes
Signed-off-by: Timo K <toger5@hotmail.de>
* device permissions
Signed-off-by: Timo K <toger5@hotmail.de>
---------
Signed-off-by: Timo K <toger5@hotmail.de>
