1187738 - WebRTC - Add label to audio and video tracks

Status: RESOLVED DUPLICATE of bug 866513

(Core :: WebRTC, defect, P2)

Description

[cjansss]

User Agent: Mozilla/5.0 (Windows NT 6.3; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/44.0.2403.89 Safari/537.36

Steps to reproduce:

0. Enable media.navigator.streams.fake in about:config
1. Browse to https://webrtc.github.io/samples/src/content/getusermedia/gum/
2. Open the console and type "streams.getVideoTracks()[0].label" and streams.getAudioTracks()[0].label


Actual results:

Empty string


Expected results:

Fake Video and Fake Audio.

This would be very useful when testing the WebRTC JS api using Selenium since it's the only way to deduce what device is actually attached to a mediaStream.

Comment 1

[Maire Reavy [:mreavy]]

We'd happily take a patch for this, but given our other priorities (and available devs), I need to put this in the parking-lot.

Comment 3

[Jan-Ivar Bruaroey [:jib] (needinfo? me)]

Label is not yet implemented in MediaStreamTrack, is the issue here, so the fake devices are not at fault, and this is a spec compliance issue, so I recommend upping priority a bit.

Comment 4

[AjayG]

We need a way to identify the source of MediaStream/Track, especially when we are sharing a video stream.
we need the attached source details like webcam/screen_number/window_name/source_process_id.

Comment 5

[Maire Reavy [:mreavy]]

Given that this is a spec compliance issue and some handy use cases are starting to emerge that need this, I'm taking this out of the parking lot and making it a low P2.  

Again, if anyone wants to contribute a patch, I'll happily make sure it gets reviewed quickly.

Comment 6

[Jan-Ivar Bruaroey [:jib] (needinfo? me)]

(In reply to AjayG from comment #4)
> we need the attached source details like webcam/screen_number/window_name/source_process_id.

Labels are meant for display, so source_process_id seems a bit too technical, but the other ones yes.

The Media Capture and Streams spec (see URL) say we "MUST return the label of the object's corresponding source, if any. If the corresponding source has or had no label, the attribute MUST instead return the empty string".

But the latest Screen Capture spec as of this writing [1] says "Each potential source of capture is treated by this API as a discrete media source". So it would follow that you'd get descriptive labels for each capture.

[1] http://w3c.github.io/mediacapture-screen-share/#device-identifiers