--- title: RTCSessionDescription slug: Web/API/RTCSessionDescription tags: - API - Audio - Experimental - Interface - Media - NeedsTranslation - Reference - TopicStub - Video - Web - WebRTC translation_of: Web/API/RTCSessionDescription ---
{{APIRef("WebRTC")}}{{SeeCompatTable}}
The RTCSessionDescription
interface describes one end of a connection—or potential connection—and how it's configured. Each RTCSessionDescription
consists of a description {{domxref("RTCSessionDescription.type", "type")}} indicating which part of the offer/answer negotiation process it describes and of the {{Glossary("SDP")}} descriptor of the session.
The process of negotiating a connection between two peers involves exchanging RTCSessionDescription
objects back and forth, with each description suggesting one combination of connection configuration options that the sender of the description supports. Once the two peers agree upon a configuration for the connection, negotiation is complete.
The RTCSessionDescription
interface doesn't inherit any properties.
{{anch("RTCSdpType")}}
describing the session description's type.This enum defines strings that describe the current state of the session description, as used in the {{domxref("RTCSessionDescription.type", "type")}} property. The session description's type will be specified using one of these values.
Value | Description |
---|---|
answer |
The SDP contained in the {{domxref("RTCSessionDescription.sdp", "sdp")}} property is the definitive choice in the exchange. In other words, this session description describes the agreed-upon configuration, and is being sent to finalize negotiation. |
offer |
The session description object describes the initial proposal in an offer/answer exchange. The session negotiation process begins with an offer being sent from the caller to the callee. |
pranswer |
The session description object describes a provisional answer; that is, a response to a previous offer that is not the final answer. It is usually employed by legacy hardware. |
rollback |
This special type with an empty session description is used to roll back to the previous stable state. |
The RTCSessionDescription
doesn't inherit any methods.
RTCSessionDescription
. The parameter is a RTCSessionDescriptionInit
dictionary containing the values to assign the two properties.signalingChannel.onmessage = function (evt) { if (!pc) start(false); var message = JSON.parse(evt.data); if (message.sdp) pc.setRemoteDescription(new RTCSessionDescription(message), function () { // if we received an offer, we need to answer if (pc.remoteDescription.type == "offer") pc.createAnswer(localDescCreated, logError); }, logError); else pc.addIceCandidate(new RTCIceCandidate(message.candidate), function () {}, logError); };
Specification | Status | Comment |
---|---|---|
{{ SpecName('WebRTC 1.0', '#rtcsessiondescription-class', 'RTCSessionDescription') }} | {{Spec2('WebRTC 1.0')}} | Initial definition. |
{{Compat("api.RTCSessionDescription")}}