Make a request, gunzip with node-compress and parse to dom with node-htmlparser.

gunzip-htmlparser-test.js

var htmlparser = require('htmlparser'),
    sys = require('sys'),
    http = require('http'),
    compress_lib = require('compress'),
    gunzip = new compress_lib.GunzipStream();

var domain = 'www.example.com'; //TODO: CHANGE ME

var rawHtml = "Xyz <script language= javascript>var foo = '<<bar>>';< /  script><!--<!-- Waah! -- -->";
var handler = new htmlparser.DefaultHandler(function (error, dom) {
    if (error) {
        sys.debug('err: ' + error);
    } else {
        sys.debug('dom: ' + JSON.stringify(dom));
    }
},
{ verbose: false, ignoreWhitespace: true });

var parser = new htmlparser.Parser(handler);
var proxy = http.createClient(80, domain);
proxy_req = proxy.request('GET', '/', {'User-Agent': 'htmlparser-test/0.0.1','Host':domain, 'Accept-Encoding':'gzip,deflate'});

gunzip.on('data', function(chunk) {
    parser.parseChunk(chunk);
}).on('end', function() {
    sys.debug('ended');
    parser.done();
});

proxy_req.on('response', function(proxy_res) {
    proxy_res.on('data', function(chunk) {
        gunzip.write(chunk);
    }).on('end', function() {
        proxy_res.connection.end();
        gunzip.close();
    });
});
proxy_req.end();