capybara-webkit

Development has been suspended on this project because QtWebKit was deprecated in favor of QtWebEngine, which is not a suitable replacement for our purposes.

We instead recommend using the Selenium or Apparition drivers.

capybara-webkit depends on a WebKit implementation from Qt, a cross-platform development toolkit. You'll need to download the Qt libraries to build and install the gem. You can find instructions for downloading and installing Qt on the capybara-webkit wiki. capybara-webkit requires Qt version 4.8 or greater.

If you're like us, you'll be using capybara-webkit on CI.

On Linux platforms, capybara-webkit requires an X server to run, although it doesn't create any visible windows. Xvfb works fine for this. You can setup Xvfb yourself and set a DISPLAY variable, try out the headless gem, or use the xvfb-run utility as follows:

xvfb-run -a bundle exec rspec

This automatically sets up a virtual X server on a free server number.

Add the capybara-webkit gem to your Gemfile:

gem "capybara-webkit"

Set your Capybara Javascript driver to webkit:

Capybara.javascript_driver = :webkit

In cucumber, tag scenarios with @javascript to run them using a headless WebKit browser.

In RSpec, use the :js => true flag. See the capybara documentation for more information about using capybara with RSpec.

Take note of the transactional fixtures section of the capybara README.

If you're using capybara-webkit with Sinatra, don't forget to set

Capybara.app = MySinatraApp.new

You can configure global options using Capybara::Webkit.configure:

Capybara::Webkit.configure do |config|
  # Enable debug mode. Prints a log of everything the driver is doing.
  config.debug = true

  # By default, requests to outside domains (anything besides localhost) will
  # result in a warning. Several methods allow you to change this behavior.

  # Silently return an empty 200 response for any requests to unknown URLs.
  config.block_unknown_urls

  # Allow pages to make requests to any URL without issuing a warning.
  config.allow_unknown_urls

  # Allow a specific domain without issuing a warning.
  config.allow_url("example.com")

  # Allow a specific URL and path without issuing a warning.
  config.allow_url("example.com/some/path")

  # Wildcards are allowed in URL expressions.
  config.allow_url("*.example.com")

  # Silently return an empty 200 response for any requests to the given URL.
  config.block_url("example.com")

  # Timeout if requests take longer than 5 seconds
  config.timeout = 5

  # Don't raise errors when SSL certificates can't be validated
  config.ignore_ssl_errors

  # Don't load images
  config.skip_image_loading

  # Use a proxy
  config.use_proxy(
    host: "example.com",
    port: 1234,
    user: "proxy",
    pass: "secret"
  )

  # Raise JavaScript errors as exceptions
  config.raise_javascript_errors = true
end

These options will take effect for all future sessions and only need to be set once. It's recommended that you configure these in your spec_helper.rb or test_helper.rb rather than a before or setup block.

The offline application cache needs a directory to write to for the cached files. Capybara-webkit will look at if the working directory has a tmp directory and when it exists offline application cache will be enabled.

capybara-webkit supports a few methods that are not part of the standard capybara API. You can access these by calling driver on the capybara session. When using the DSL, that will look like page.driver.method_name.

console_messages: returns an array of messages printed using console.log

// In Javascript:
console.log("hello")

# In Ruby:
page.driver.console_messages
=> [{:source=>"http://example.com", :line_number=>1, :message=>"hello"}]

error_messages: returns an array of Javascript errors that occurred

page.driver.error_messages
=> [{:source=>"http://example.com", :line_number=>1, :message=>"SyntaxError: Parse error"}]

cookies: allows read-only access of cookies for the current session

page.driver.cookies["alpha"]
=> "abc"

header: set the given HTTP header for subsequent requests

page.driver.header 'Referer', 'https://www.thoughtbot.com'

capybara-webkit is Copyright (c) 2010-2015 thoughtbot, inc. It is free software, and may be redistributed under the terms specified in the LICENSE file.

Thank you, contributors!

Code for rendering the current webpage to a PNG is borrowed from Phantom.js' implementation.

capybara-webkit is maintained and funded by thoughtbot, inc. The names and logos for thoughtbot are trademarks of thoughtbot, inc.

We love open source software! See our other projects or hire us to help build your product.