Collage brings together many public APIs, along with a method of presenting media within a limitless two-dimesional space to create memorable and meaningful experiences. It's a mixture between Big Surface, Giant Quadtree, and a public API library.

For plain JavaScript applications, use the dist/collage.js, which will inject the Collage module object to the global scope (i.e. it creates window.collage). You can also use this file as an AMD or CommonJS module. If you use bower, you can install collage with bower install collage.

Depending on which loaders you use, you will need to include these in your HTML doc, before starting the collage:

<!-- Youtube Iframe API -->
<script src="//www.youtube.com/iframe_api"></script>

<!-- Twitter API -->
<script src="//platform.twitter.com/widgets.js"></script>

<!-- Facebook API -->
<div id="fb-root"></div>
<script>
	(function(d, s, id) {
	  var js, fjs = d.getElementsByTagName(s)[0];
	  if (d.getElementById(id)) return;
	  js = d.createElement(s); js.id = id;
	  js.src = "//connect.facebook.net/en_US/all.js#xfbml=1";
	  fjs.parentNode.insertBefore(js, fjs);
	}(document, 'script', 'facebook-jssdk'));
</script>

These are referred to as loaders in the codebase.

• Facebook (like boxes)
• Reddit (images and embeds)
• Youtube
• Flickr
• Google News
• Google Plus
• Twitter (currently broken due to auth changes)
• NY Times (requires a reverse-proxy due to lack of CORS headers)

Check the loader configurations for how to query them for content.

Collage extends Big Surface, all API methods that are available for Big Surface is also available for Collage. In the API collage refers to the module object and collageInstance refers to a Collage instance which is created via collage.create.

Constructor function which returns a new Collage instance. The container is the element which will contain the collage.

Loads the media from the given loaderName according to the loaderConfig and assigns the loaded media to the tagName.

var collageInstance = collage.create(document.body);
collageInstance.load('some fb content', 'facebook', {
    minLikes: 200,
    query: 'popcorn'
}).then(function(){
	// Wait for the loading to be finished, then
	// start the collage (filling the center) and set the active tags
	collageInstance.start('some fb content');
	collageInstance.speed(8);
});

Loads the media from the given loaderName : loaderConfig mapping and assigns the loaded media to the tagName.

var collageInstance = collage.create(document.body);

collageInstance.load('my content', {
  facebook: [
    {    
      minLikes: 200,
      query: 'popcorn'
    },
    {
      ids: ["110452248983275"]
    }
  ],
  twitter: [
    { query: 'popcorn' },
    { query: 'caramel' }
  ]
}).then(function(){
	// Wait for the loading to be finished, then
	// start the collage (filling the center) and set the active tags
	collageInstance.start('my content');
	collageInstance.speed(8);
});

Starts the transform loop of the collage (allows for movement) and optionally sets the active tags to the given arguments.

Brings the collage movement to a halt within the given time duration (or immediately if none provided).

Resumes collage movement to the state it was in previous to a pause call. If a duration is provided, the speed will tween to the limit within the given time period.

Sets the given tags as the 'active' tags which media for the collage will be pulled from.

Adds the given element(s) to the given tag(s).

Removes the given element(s) from the given tag(s).

Gets the elements for the given tag names.

Inserts an element directly in the collage surface at left, top (in pixels). If show is set to true, the element is placed visibly, otherwise hidden.

A map of available loaders which you can extend during run-time.

Fills the visible center of the screen with elements.

Loaders are where you get the content for the collage from. The content from the loaders are associated with one or more tags, which the collage will use to populate itself with.

Adds like boxes for facebook pages to the collage

• query: a search query (not required if providing ids)
• ids: an array of page ids to load (not required if providing a query)
• limit (default 3): max number of results from query search
• width (default 400): width of the like box
• height (default 600): height of the like box
• minLikes (default 0): minimum number of likes for results of the query search
• showFaces (default true): whether to show faces in the like box
• showStream (default true): whether to show the page stream in the like box
• showHeader (default false): whether to show a 'like our page' header in the like box

var collageInstance = collage.create(document.body);
collageInstance.load('some fb content', 'facebook', {
    minLikes: 200,
    query: 'one man one woman'
}).then(function(){ // wait for the first batch to complete loading
	collageInstance.load('more fb content', {
	  facebook: [{    
	      minLikes: 200,
	      query: 'popcorn'
	   },
	   {
	      ids: ["110452248983275"]
	   }]
	}).then(function(){
		// Wait for the loading to be finished, then
		// start the collage (filling the center) and set the active tags
		collageInstance.start('some fb content', 'more fb content');
		collageInstance.speed(8);
	});	
});

Adds images or reddit embeds to the collage

• type default "photos", can be "embed" or "photos"
• sort default "top", can be "confidence", "top", "new", "hot", "controversial", "old", "random"
• subreddit default "all"
• limit default 20
• time default "all", can be "hour", "day", "week", "month", "year", "all"
• nsfw default "false"
• minComments default 0
• minScore default 0
• width default 500, width of the embed box
• height default 600, height of the embed box

collageInstance.load('reddit popcorn', {
reddit: [
  {
		query: "popcorn site:imgur.com subreddit:funny",
		limit: "10"
  },
  {
  	type: "embed",
  	query: "popcorn"
  }]
}).then(function(){
	// Wait for the loading to be finished, then
	// start the collage (filling the center) and set the active tags
	collageInstance.start('reddit popcorn');
	collageInstance.speed(8);        
});

Choose one:

• query a search query for tweets
• ids an array of tweet ids
• id a single tweet id

collageInstance.load('popcorn', {
 twitter: [
  	{ query: "popcorn"}
  ],
}).then(function(){
	// Wait for the loading to be finished, then
	// start the collage (filling the center) and set the active tags
	collageInstance.start('popcorn');
	collageInstance.speed(8);        
});

Adds youtube videos to the collage

Choose one:

• query a string to search Youtube with
• videoId a singe video id to load
• videoIds an array of video ids to load

Optional parameters:

• mute default false
• loop default false
• autoplay default false
• callback
• continuousPlay default false. This option determines if the video pauses when it goes out of view or not.
• startTime default 0
• duration default "short"
• key Google api api key
• width default 1060
• height default 650

collageInstance.load('popcorn', {
  youtube: [
    {
  		query: "popcorn",
  		mute: true,
  		loop: true
  	},
    {
  		videoId: 'abcdefgh',
  		mute: true,
  		autoplay: true,
  		loop: true,
  		startTime: 108
  	}
  ]
}).then(function(){
	// Wait for the loading to be finished, then
	// start the collage (filling the center) and set the active tags
	collageInstance.start('popcorn');
	collageInstance.speed(8);        
});

Adds images from Flickr to the collage

• tags the search query for tags to load
• sort default "relevance"
• count default 20
• license default "1,2,3,4,5,6,7,8" see Flickr API
• apiKey
• tagMode default "all"
• isCommons default false
• contentType default 1 (photos)

collageInstance.load('popcorn', {
  flickr: [
    {  
			tags: "popcorn",
			sort: "interestingness-desc"
		}
  ],
}).then(function(){
	// Wait for the loading to be finished, then
	// start the collage (filling the center) and set the active tags
	collageInstance.start('popcorn');
	collageInstance.speed(8);        
});

Adds news articles from Google News to the collage.

The loader only accepts a string query.

collageInstance.load('popcorn', {
  googleNews: ["popcorn"]
}).then(function(){
	// Wait for the loading to be finished, then
	// start the collage (filling the center) and set the active tags
	collageInstance.start('popcorn');
	collageInstance.speed(8);        
});

Adds posts from Google Plus to the collage.

The loader only accepts a string query.

collageInstance.load('popcorn', {
  googlePlus: ["popcorn"]
}).then(function(){
	// Wait for the loading to be finished, then
	// start the collage (filling the center) and set the active tags
	collageInstance.start('popcorn');
	collageInstance.speed(8);        
});

Adds news articles from the NY Times API. The api endpoint must be proxied locally, since it doesn't allow for JSONP calls to be made.

Choose one:

• query A live query to make to the NY Time api
• data A pre-recorded JSON object to use instead of doing a query

collageInstance.load('popcorn', {
	nyTimes: [{
		query: "popcorn+nytd_des_facet%3A%5BHate+Crimes%5D"
	}]
}).then(function(){
	// Wait for the loading to be finished, then
	// start the collage (filling the center) and set the active tags
	collageInstance.start('popcorn');
	collageInstance.speed(8);        
});

Adds an iframe to the collage. Becareful, this is a particularly expensive thing to add, because it embed an entire site in the collage.

• url
• width
• height

collageInstance.load('popcorn', {
  		iframe: [{
				url: "https://en.wikipedia.org/wiki/Popcorn",
				width: 800,
				height: 600
			}]
}).then(function(){
	// Wait for the loading to be finished, then
	// start the collage (filling the center) and set the active tags
	collageInstance.start('popcorn');
	collageInstance.speed(8);        
});

Adds a single image to the collage.

The loader only accepts a string src.

collageInstance.load('popcorn', {
  image: ["https://www.popcorn.com/popcorn.jpg"]
}).then(function(){
	// Wait for the loading to be finished, then
	// start the collage (filling the center) and set the active tags
	collageInstance.start('popcorn');
	collageInstance.speed(8);        
});

This project uses boilerplate-gulp. Run 'gulp dev' to develop and have incremental builds, continuous testing, etc. Run 'gulp' to run regenerate the dist files.