techno.blog("Dion")

A core vision of Gears is that the same APIs are available on the desktop, and mobile versions.

That being said, there are APIs that will potentially make more sense on one platform or another. One of those that has garnered excitement on the mobile platform is the Location API.

Another is the newly proposed Camera API that “enables a script to trigger the camera viewfinder and allow a picture or video to be captured using the device’s onboard camera.”

That being said, this could still tie into a computers webcam.

Summary

The goal of this API is to allow a Web application to launch the native camera application and obtain the resulting media file.

On devices with built-in camera, all of the UI interaction is left to the corresponding native camera application, not Gears.

On devices that do not have a built-in camera (e.g. laptop or desktop computers), Gears will scan for attached Web cameras and will provide a basic viewport that allows the users to grab a picture or record a movie. If several cameras are attached, Gears will provide a way for the users to select the device they want to use.

This API is based on the “action-driven permission model” - the user is involved in the action and hence a separate permission prompt is not required.

You would use the API like this:

// Create a camera object
var camera = google.gears.factory.create("beta.camera");
 
// Capture an image.
var image = camera.captureImage(function(media) {
    uploadImage(media.content);
});
 
// Capture of a video, specifying a desired resolution and duration.
var video = camera.captureVideo(function(media) {
    uploadVideo(media.content);
}, 5000);

The full API proposal is:

interface Camera {
  // Acquisition methods
 
  // Triggers the camera viewfinder and invokes the callback after the user
  // has either captured the image or canceled the operation.
  void captureImage(CaptureCallback callback, optional CaptureOptions options);
  // Triggers the camera viewfinder and invokes the callback after the user
  // has either captured the video, canceled the operation or maxDuration milliseconds
  // have elapsed.
  void captureVideo(CaptureCallback callback, double maxDuration, optional CaptureOptions options);
 
  // Query methods
 
  Array availableImageFormats();   
  Array availableVideoFormats();   
};
 
void CaptureCallback(Media media);
 
interface CaptureOptions {
  string format;  // MIME type
  double width;   // desired image or video frame width
  double height;  // desired image or video frame height
};
 
interface Media {
  readonly string format;  // MIME type
  readonly double width;   
  readonly double height;
  readonly Blob content;   // reference to captured media
};

What do you think? What haven’t we thought about yet?

Other Future APIs

• Resumable Uploads via PUT/POST
• Open ID and OAuth
• Services / Daemon API
• Notification API
• Crypto API
• Developer Tools
• Factory API updates
• Blob API
• Logging API
• Messaging API
• Location API
• Desktop Shortcut API
• Image Manipulation API
• Introduction to the series

Disclaimer: These APIs are being discussed by the Gears community and do not represent live products. Do you have ideas for cool Gears that make the Web better? Let us know!.