quick and dirty WScript email crawler usage: from command prompt: "cscript foo.js"

emailspider.js

var url = "http://thenewobjective.com/blog/",
    email = /\b[A-Z0-9+_.-]+@(?:[A-Z0-9-]+\.)+[A-Z]{2,6}\b/gi,
    delay = 2000,
    visited = {}, hostname,
    baseDocument = new ActiveXObject("htmlfile"),
    a = baseDocument.createElement("a");

a.href = url;
hostname = a.hostname;

Array.prototype.undupe = function(){
    var seen = {}, result = [];
	for (var i = 0, len = this.length; i < len; ++i){
		var val = this[i];
		if (!seen[val]) {
			seen[val] = 1;
			result.push(val);
		}
	}
	return result;
}

function getAddresses(url, results){
    a.href = url;
    a.href = url.replace(a.hash,"");

    if(a.hostname !== hostname){
        WScript.Echo("Skipping: " + a.href);
        return results;
    }

    if(visited[a.href]){
        WScript.Echo("Already Visited: " + a.href);
        return results;
    }

    WScript.Echo("Parsing: " + a.href);
    visited[a.href] = true;

    try {
        WScript.Echo("Waiting");
        WScript.Sleep(delay);
        var document = WScript.GetObject(a.href);

        while(document.readyState !== "complete")
            WScript.Sleep(200);

		var emails = document.documentElement.innerHTML.match(email) || [];
		
		WScript.Echo(emails.length + " addresses found" );
		
		for(var j = 0, len = emails.length; j < len; j++)
			results[results.length] = emails[j];

        var links = document.getElementsByTagName("a");

        WScript.Echo(links.length + " links on page");

        for (var i = 0, len = links.length; i < len; i++)
            getAddresses(links[i].href,results)
    } catch(e) {
        WScript.Echo("Unable to parse: " + url);
		WScript.Echo(e.message);
    } finally{
		return results
	}
}

var res = getAddresses(url, []).undupe();

WScript.Echo("done. " + res.length + " total addresses found:");
WScript.Echo("-------------------------")
WScript.Echo(res.join("\n"));