PhantomJs Escaped Fragment Crawler

PhantomJsCrawl.js

var fs = require('fs'),
    //root url to start crawl from, it will only look for hashes in a links for the specified baseUrl
    //so for instance, if you had a link to some other website http://www.abc.com/something-here it's NOT going to parse it
    baseUrl = 'http://localhost:3000/',
    //folder to save the html pages to
    saveFolder = 'public/_escaped_fragment_/',
    //array containing every link already parsed
    parsedLinks = [];
//method to parse given url/hash and iterator that recursively calls the additional pages
var checkPage = function (page, url, hash) {
    parsedLinks.push(url + hash);
    page.open(url + hash, function (status) {
        var filePath = hash.replace('#!', '').replace('#', '') + '.html';
        if (filePath === '.html') {
            filePath = 'Index.html';
        }
        console.log(filePath);
        setTimeout(function (){
            var content = page.evaluate(function () {
                return document.body.innerHTML;
            });
            fs.write(saveFolder + filePath, content, 'w');
            var hashes = page.evaluate(function () {
                var elements = document.getElementsByTagName('a');
                hashes = [];
                for (var i = 0; i < elements.length; i++) {
                    if (elements[i].hash !== '' && elements[i].href.indexOf(location.href) === 0) {
                        hashes.push(elements[i].hash);
                    }
                }
                return hashes
            });
            
            hashes.forEach(function (hash) {
                if (parsedLinks.indexOf(url + hash) == -1) {
                    checkPage(require('webpage').create(), url, hash);
                }
            });
        }, 2000)
    });
};
//start the crawl
checkPage(require('webpage').create(), baseUrl, '');