browser

browser

[Node.js] browsing urls with cookies, that is, we can scrape with authenticated pages!

Installation

git clone git://github.com/shinout/browser.git

OR

npm install browser 

Features

• automatic cookie management
• easy asynchronous handling with Junjo.js

Usage

helloworld (onetime access)

var browser = require("browser");
browser.browse("shinout.net", function(err, out) {
  console.log(out.result);
});

helloworld2 (using object)

var browser = require("browser");
var $b = new browser();
$b.browse('https://accounts.google.com/Login'); // browse this url
$b.run(); // execution

login sample (requires jquery)

var userdata = {
  email: "[e-mail]",
  pass : "XXXXXXXX"
};

var $b = new browser();
$b.submit({
  from : 'https://accounts.google.com/Login',
  selector: "#gaia_loginform",
  data : {
    Email  : userdata.email,
    Passwd : userdata.pass
  }
});

// authenticated access
$b.browse('https://mail.google.com/mail/u/0/?ui=html&zy=d')
.after(); // browse after previously registered function

login sample2 (do what $b.submit() is doing manually)

// $b.browse(the label of this request, url to access)
$b.browse('login', 'https://accounts.google.com/Login');

/* $b.browse(function(
 *   err : errors occured in the previous request, 
 *   out : result of the previous browsing
 *) { return url or return [url, options] }
 */
$b.browse(function(err, out) {
  var window = jsdom(out.result).createWindow();
  var $ = jquery.create(window);
  var postdata = {
    Email  : userdata.email,
    Passwd : userdata.pass
  };
  var url = $("#gaia_loginform").attr("action");
  // get hidden fields, and register them to post data
  $("input").each(function(k, el) {
    var $el = $(el);
    var name = $el.attr("name"), type = $el.attr("type"), val = $el.val();
    if (type == "hidden" || type == "submit") postdata[name] = val;
  });
  return [url, {
    data  : postdata, // set post data
    method: "POST"    // set HTTP method (default: GET)
  }];
})
.after("login"); // browse after browsing with label="login"

$b.browse('https://mail.google.com/mail/u/0/?ui=html&zy=d')
.after(); // browse after previously registered function

/* running on end of all browsings
 *   err: error object or null
 *   out: { result : result body, ...}
 */
$b.on("end", function(err, out) {
  if (!out || !out.result) return res.end("no result");
  res.writeHead(200, {'Content-Type' : 'text/html'});
  console.log("result", out.result);
  res.end(out.result);
});
$b.run();

options object

option object to pass to $b.browse() is the same format as u2r options. See u2r in detail. The following are common options.

• data : (object) key-value pairs to pass to server
• method : HTTP method (GET|POST|PUT|DELETE|HEAD). default: GET

all other values below are automatically generated from URL

• host
• protocol
• path
• port
• body : querystring format of options.data

keys of out object

• result : response data(Buffer or String)
• statusCode
• location
• responseHeaders
• cookies : set-cookie headers
• url : browsed url

Contact

Feel free to contact @shinout!