abergmeier/chunked.js

Polyfill for Streams API

chunked.js

Polyfill for chunked XMLHttpRequest extension proposed by Mozilla.

Was started as polyfill for Streams API, but due to the sheer size of the spec, it was reduced to handle just downloading, which is exactly, what Mozilla's extension does.

###Scenario: The main usage for this should be with downloading large files via http. It aims to fix problems like:

• Large files need a long time to download. Currently processing them is only possible after loading has finished. With chunked.js you can start processing as soon as the first chunk (scnr) was received.
• Large files may result in an Javascript heap overflow. Via chunked.js you can choose to only process a smaller fraction at a time.

###Usage:

XMLHttpRequest.responseType = < normal types > | "chunked-text" | "chunked-blob" | "chunked-arraybuffer";

Setting one of the chunked response types, results in XMLHttpRequest.response being only available/set in progress events (in contrast to load event). XMLHttpRequest.response then contains the data received since the last progress event.

###Example: <script src="chunked.min.js"></script>

var req = new XMLHttpRequest();
req.open( "GET", "http://ipv4.download.thinkbroadband.com/512MB.zip" );
req.responseType = "chunked-array" // alternate `chunked-text` and `chunked-blob`
req.onload = function( event ) {
};
req.onprogress = function( event ) {
    // Access chunk, between last progress call and now
    var chunk = req.result;
    my_incredible_function( chunk );
};
req.send();

###Internas: Wrapper is not derived (prototype = new XMLHttpRequest()) from XMLHttpRequest, since Javascript forbids this for DOMElements. It wraps XMLHttpRequest as well as possible. Be aware that naive prototype-checking may fail, though! Does not yet work for ftp!