Asset loading At a glance

For some types of browser applications, namely video games and other media-rich applications, it is necessary to request batches of assets to be fetched programmatically. Often times this is done before caching them in memory, localStorage, or a webDB to be consumed syncronously by your application.

WEBSERVER <------> BROWSER -> CACHE?

Why an asset loader?

It is absolutely true that each asset could simply be requested individually and then processed. However, the semantics required by a web-browser to fetch files from a remote machine differ by file type. IE. Audio objects, Image objects, JSON data, even other javascripts are all fetched slightly differently.

The purpose of this loader is to allow the details of that fetching to remain opaque to the consumer while providing a single higher-level API for fetching a wide variety of assets.

Supported data types

The asset loader supports several data types out of the box AND has a simple api for overriding these built-ins OR for adding additional fetching routines.

Loader.Asset (url, type, name)

This is a constructor function used to define assets prior to fetching. All parameters are required and are strings

Loader.addType (typeName, fetchFn)

Add a handler for a specified type. N.B. Type is an arbitrary distinction specified by the consumer of this library. The provided fetchFn should return a promise for the eventual LoadedAsset.

Loader.removeType (typeName)

Remove a handler for a specified type.

Loader.getType (typeName)

Return the fetch function for a specified type

Loader.fetch (Asset)

Fetch a single asset and return a promise for a Loaded Asset.

Loader.fetchMany (Assets)

Fetch many assets and callback when all have either succeeded/failed. The param to the resolve handler of the promise will be an array of objects. Each object has an attribute "succeeded" which is true/false and an attribute "result" which is either a LoadedAsset or the origin Asset.