Collecting every word in document DOM tree with javascript

Question

Suppose you have a large document with around ~7000 words. I need to send all data to server. I have no chance to use jquery, prototype etc. It should be clean OO javascript. Sample page would be json russian page I will exclude all tags and html markup from words.

My question is;

1. How can i collect/harvest all (utf8) words from document?

2. How can i convert the result to JSON data?

Thanks

Answer 1

This really doesn't seem like a job for Object Oriented programming. A sexy recursive function would work much better.

 var output=[];

 function scan(element) {
    var children=element.childNodes;
    for (var child in children){
        if (children[child].nodeValue) {
            output.push(children[child].nodeValue);
        }else {
            scan(children[child]);
        };
    };
 };

 scan(window.document.body);

This doesn't break the text up into individual words or even produce JSON, but it will give you a list of the individual words. You still need to do some cleanup on the text. In my 2 seconds of testing I found that it likes to display the text of everything including javascript and newlines (\n). Maybe if I feel like it I'll add more code. But this should get you going.

For turning it into JSON try Douglas Crockford's toJSON code. Just google it.