媒体设备入门

When developing for the web, the WebRTC standard provides APIs for accessing cameras and microphones connected to the computer or smartphone. These devices are commonly referred to as Media Devices and can be accessed with JavaScript through the navigator.mediaDevices object, which implements the MediaDevices interface. From this object we can enumerate all connected devices, listen for device changes (when a device is connected or disconnected), and open a device to retrieve a Media Stream (see below).

The most common way this is used is through the function getUserMedia(), which returns a promise that will resolve to a MediaStream for the matching media devices. This function takes a single MediaStreamConstraints object that specifies the requirements that we have. For instance, to simply open the default microphone and camera, we would do the following.

const constraints = {
    'video': true,
    'audio': true
}
navigator.mediaDevices.getUserMedia(constraints)
    .then(stream => {
        console.log('Got MediaStream:', stream);
    })
    .catch(error => {
        console.error('Error accessing media devices.', error);
    });

const openMediaDevices = async (constraints) => {
    return await navigator.mediaDevices.getUserMedia(constraints);
}

try {
    const stream = openMediaDevices({'video':true,'audio':true});
    console.log('Got MediaStream:', stream);
} catch(error) {
    console.error('Error accessing media devices.', error);
}

The call to getUserMedia() will trigger a permissions request. If the user accepts the permission, the promise is resolved with a MediaStream containing one video and one audio track. If the permission is denied, a PermissionDeniedError is thrown. In case there are no matching devices connected, a NotFoundError will be thrown.

The full API reference for the MediaDevices interface is available at MDN web docs.

查询媒体设备

In a more complex application, we will most likely want to check all the connected cameras and microphones and provide the appropriate feedback to the user. This can be done by calling the function enumerateDevices(). This will return a promise that resolves to an array of MediaDevicesInfo that describe each known media device. We can use this to present a UI to the user which let's them pick the one they prefer. Each MediaDevicesInfo contains a property named kind with the value audioinput, audiooutput or videoinput, indicating what type of media device it is.

function getConnectedDevices(type, callback) {
    navigator.mediaDevices.enumerateDevices()
        .then(devices => {
            const filtered = devices.filter(device => device.kind === type);
            callback(filtered);
        });
}

getConnectedDevices('videoinput', cameras => console.log('Cameras found', cameras));

async function getConnectedDevices(type) {
    const devices = await navigator.mediaDevices.enumerateDevices();
    return devices.filter(device => device.kind === type)
}

const videoCameras = getConnectedDevices('videoinput');
console.log('Cameras found:', videoCameras);

监听设备变化

Most computers support plugging in various devices during runtime. It could be a webcam connected by USB, a Bluetooth headset, or a set of external speakers. In order to properly support this, a web application should listen for the changes of media devices. This can be done by adding a listener to navigator.mediaDevices for the devicechange event.

// Updates the select element with the provided set of cameras
function updateCameraList(cameras) {
    const listElement = document.querySelector('select#availableCameras');
    listElement.innerHTML = '';
    cameras.map(camera => {
        const cameraOption = document.createElement('option');
        cameraOption.label = camera.label;
        cameraOption.value = camera.deviceId;
    }).forEach(cameraOption => listElement.add(cameraOption));
}

// Fetch an array of devices of a certain type
async function getConnectedDevices(type) {
    const devices = await navigator.mediaDevices.enumerateDevices();
    return devices.filter(device => device.kind === type)
}

// Get the initial set of cameras connected
const videoCameras = getConnectedDevices('videoinput');
updateCameraList(videoCameras);

// Listen for changes to media devices and update the list accordingly
navigator.mediaDevices.addEventListener('devicechange', event => {
    const newCameraList = getConnectedDevices('video');
    updateCameraList(newCameraList);
});

Media constraints:媒体限制

The constraints object, which must implement the MediaStreamConstraints() interface, that we pass as a parameter to getUserMedia() allows us to open a media device that matches a certain requirement. This requirement can be very loosely defined (audio and/or video), or very specific (minimum camera resolution or an exact device ID). It is recommended that applications that use the getUserMedia() API first check the existing devices and then specifies a constraint that matches the exact device using the deviceId constraint. Devices will also, if possible, be configured according to the constraints. We can enable echo cancellation on microphones or set a specific or minimum width and height of the video from the camera.

async function getConnectedDevices(type) {
    const devices = await navigator.mediaDevices.enumerateDevices();
    return devices.filter(device => device.kind === type)
}

// Open camera with at least minWidth and minHeight capabilities
async function openCamera(cameraId, minWidth, minHeight) {
    const constraints = {
        'audio': {'echoCancellation': true},
        'video': {
            'deviceId': cameraId,
            'width': {'min': minWidth},
            'height': {'min': minHeight}
            }
        }

    return await navigator.mediaDevices.getUserMedia(constraints);
}

const cameras = getConnectedDevices('videoinput');
if (cameras && cameras.length > 0) {
    // Open first available video camera with a resolution of 1280x720 pixels
    const stream = openCamera(cameras[0].deviceId, 1280, 720);
}

The full documentation for the MediaStreamConstraints interface can be found on the MDN web docs.

本地回放

Once a media device has been opened and we have a MediaStream available, we can assign it to a video or audio element to play the stream locally.

async function playVideoFromCamera() {
    try {
        const constraints = {'video': true, 'audio': true};
        const stream = await navigator.mediaDevices.getUserMedia(constraints);
        const videoElement = document.querySelector('video#localVideo');
        videoElement.srcObject = stream;
    } catch(error) {
        console.error('Error opening video camera.', error);
    }
}

The HTML needed for a typical video element used with getUserMedia() will usually have the attributes autoplay and playsinline. The autoplay attribute will cause new streams assigned to the element to play automatically. The playsinline attribute allows video to play inline, instead of only in full screen, on certain mobile browsers. It is also recommended to use controls="false" for live streams, unless the user should be able to pause them.

<html>
<head><title>Local video playback</video></head>
<body>
    <video id="localVideo" autoplay playsinline controls="false"/>
</body>
</html>