Intent to Deprecate and Remove: RTCPeerConnection#getStreamById(id)

[Philip Jägenstedt]

Summary

Remove the getStreamById() method for RTCPeerConnection, which is a convenience method for finding a local or remote stream by a particular id.

Motivation

It was removed from the spec in 2014, removed from Gecko, and appears to be used very little in the wild.

Interoperability and Compatibility Risk

The interop risk is associated with Edge and WebKit-based browsers other than Safari at this point.

Edge: No signals. I've filed an issue for removal requesting feedback.

Firefox: Removed

Safari: RTCPeerConnection is not shipped. I've filed an issue for WebKit requesting feedback.

If there is any feedback, I will report back here.

The compat risk was judged by httparchive analysis, where the total number of hits was low, and no cases that would break were found. It looks much safer than some removals we've successfully done.

Alternative implementation suggestion for web developers

From the bug, here is a replacement for anyone that needs it after removal: RTCPeerConnection.prototype.getStreamById = function(id) { try { var localStreams = this.getLocalStreams(); var remoteStreams = this.getRemoteStreams(); var i; for (i = 0; i < localStreams.length; i++) { if (localStreams[i].id == id) return localStreams[i]; } for (i = 0; i < remoteStreams.length; i++) { if (remoteStreams[i].id == id) return remoteStreams[i]; } } catch(e) {} return null; }

The try-catch is because getLocalStreams() and getRemoteStreams() are also non-standard, with an uncertain fate.

Usage information from UseCounter

https://www.chromestatus.com/metrics/feature/timeline/popularity/1645

Rounds to 0, gaps in the timeline means it was literally zero that day.

OWP launch tracking bug

https://crbug.com/698163

Entry on the feature dashboard

https://www.chromestatus.com/feature/5751819573657600

Requesting approval to remove too?

Yes, I would like to deprecate in M60 and remove in M62.

[PhistucK]

Just a note - I am not sure an HTTPArchive analysis makes sense for WebRTC cases. Homepages would rarely use WebRTC.

☆PhistucK

--
You received this message because you are subscribed to the Google Groups "blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+unsubscribe@chromium.org.

[TAMURA, Kent]

LGTM1

--

TAMURA Kent
Software Engineer, Google

[Dimitri Glazkov]

LGTM2

[Chris Harrelson]

LGTM3

LGTM2

[Philip Jägenstedt]

This has now been removed.

To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+...@chromium.org.