Re: [webrtc-pc] Default ufrag for addIceCandidate when there are different ufrags in SDP

`addIceCandidate` requires an `sdpMid` or `sdpMLineIndex`, which identifies which media stream the candidate is being added for. So if `ufrag` is absent, the candidate is assumed to belong to the last ICE generation used by that "m=" section.

For example, if SRD is called with:

```
m=video ...
a=mid:video
a=ice-ufrag:video_ufrag_1
...
m=audio ...
a=mid:audio
a=ice-ufrag:audio_ufrag_1
```

And later with:

```
m=video ...
a=mid:video
a=ice-ufrag:video_ufrag_2
...
m=audio ...
a=mid:audio
a=ice-ufrag:audio_ufrag_2
```

Then if `addIceCandidate` is called with `{sdpMid: "audio", candidate: "..."}` with no ufrag, the candidate is assumed to belong to the ICE generation identified by "audio_ufrag_2".

Does this make sense? Is there a way the spec could be reworded to make this more clear?

-- 
GitHub Notification of comment by taylor-b
Please view or discuss this issue at https://github.com/w3c/webrtc-pc/issues/1219#issuecomment-302238012 using your GitHub account