Browserjet - headless webkit with a node.js interface

Browserjet gives you access to headless webkit which is often useful for site scraping, testing, page screenshots, saving pdfs, and more.

Installing

Browserjet's binary depends on Qt4 and Qt-Webkit, therefore you will need it installed on your system for the build to succeed. Only Linux 32/64 bits have been tested so far.

Ubuntu: apt-get install libqtwebkit-dev

Once dependencies are satisfied:

npm install browserjet

Usage

Browserjet provides and easy to use API. To load, manipulate, and save a page as a png you can use:

var browserjet = require('browserjet'),
    browser = browserjet.createBrowser();
    
browser
  .chain
  .get("http://nodejs.org/")
  .run(function() {
    document.body.style.backgroundColor = '#000066';
  })
  .save('nodejs.png')
  .end(function(err) {
    if(err) { throw err; }
  });

API

browser.get(url, callback)

Fetch the given page and pass ok to callback after completely loaded

browser.get("http://google.com/", function(err, ok) {
  if(err) { throw err; }
  console.log("Page loaded");
});

Also see examples/loadspeed.js for another example

browser.setHtml(html, callback)

Set the browsers content to the given HTML and run callback when completely loaded

browser.setHtml("<html><body>Hello World</body></html>");

browser.html(callback)

Return the browsers HTML

browser.run(js, callback)

Run the supplied javascript synchronously in the context of the currently loaded page and pass the last result to the callback

browser.run("document.getElementsByTagName('p').length", function(err, count) {
  console.log("There are "+count+" paragraphs in the currently loaded page");
});

browser.runAsync(js, callback)

An advanced command to run the given javascript asynchronously in the context of the currently loaded page, browserjet.callback is exposed on the page and returns its parameter to the callback.

browser.runAsync(function() {
  // This code runs in the context of the loaded page
  setTimeout(function() {
    browserjet.callback(document.title);
  }, 1000);
}, function(err, title) {
  console.log(title);
});

After 1s, the title of the currently loaded page is printed

Also see examples/qunit.js for another example of runAsync

browser.save(filename, callback)

Saves the current page, supported file extensions are image formats (png, gif, svg, tiff) and pdf

browser.save("page.pdf");
browser.save("/tmp/page.png");

Tests

The browserjet tests use expresso, which you can install with

npm install -g expresso

Then in the module directory you can just run

expresso

ssimono/browserjet