Jeremy Lainé
Jeremy Lainé
Also of interest: https://github.com/jitsi/ice4j/blob/master/doc/configuration.md https://wiki.mozilla.org/Media/WebRTC/Privacy
Creating the datagram socket to determine the default route is a neat trick, will keep that in mind. Binding to 0.0.0.0 is something I've tried previously but I seem to...
At the moment I don't have a usecase for it, as my primary goal for `aioice` is to support `aiortc` (a Python WebRTC implementation). I'd certainly be willing to review...
The status remains the same, I don't have a usecase for it so there is little chance I'll implement it myself. I'm happy to review pull requests though.
An update here : we now support a much broader array of TURN transports, namely UDP/TCP/TLS. The API is starting to be a little heavy with a proliferation of turn_XXX...
Let's keep the discussion here. TURN or STUN slowing down the process is a symptom of the same limitation: candidates are collected in one go.
@rprata I think the unit test are you best source of information TBH, they illustrate a number of scenarii. To give you some starting points: - look at the `gather()`...
Yes the RFC you linked seems to be the correct one. Right now aioice is a "half trickle" implementation if I'm not mistaken.
I don't think this is correct. According to the RFC, the periodic timer triggers new checks whenever it fires, I don't see anything saying the checks must be performed one...