Adds a test for the integration between components.

jcxplorer · May 15, 2011 · 9b9d129 · 9b9d129
1 parent e08e713
commit 9b9d129
Showing 1 changed file with 137 additions and 0 deletions.
diff --git a/spec/acceptance/tracking_spec.rb b/spec/acceptance/tracking_spec.rb
@@ -0,0 +1,137 @@
+## Snowfinch tracking test
+#
+# [Snowfinch](https://github.com/jcxplorer/snowfinch) is a realtime web
+# analytics application written in
+# [Ruby on Rails](http://rubyonrails.org/) and using
+# [MongoDB](http://www.mongodb.org/).
+#
+# This test is an example of how to test several components of an
+# application.
+#
+#
+### Background
+#
+# The system contains three components:
+#
+# * A JavaScript tracker that executes requests containing click
+#   information.
+# * A collector application that records the click information from the
+#   requests executed by the JavaScript tracker.
+# * A Rails application where users can manage sites and see the
+#   analytics data.
+#
+# This test goes through logging in, creating a site, simulating
+# copy-pasting of the displayed tracking code into a page, loading the
+# page, and checking that the application reports one active visit,
+# one pageview, and one visitor. This are the same steps a user would
+# follow when using Snowfinch.
+#
+#
+### Why?
+#
+# Components are tested individually, having automated acceptance and
+# unit tests, but nothing ensures that the system works as a whole. This
+# test helps ensure that everything is wired up correctly to work.
+#
+#
+### Techniques
+#
+# Snowfinch is implemented using behavior driven development, but with
+# acceptance tests written in Ruby instead of plain English. This allows
+# for faster test and development cycles, and having user stories in
+# plain English would not bring any big benefit to this specific
+# application.
+#
+#
+### Tools
+# 
+# The tools used for testing are:
+#
+# * [RSpec](https://github.com/rspec/rspec) is a testing suite for
+#   behavior driven development, complete with a mocking framework and
+#   Rails integration.
+# * [Capybara](https://github.com/jnicklas/capybara) is an acceptance
+#   test framework for web applications.
+# * [Steak](https://github.com/cavalle/steak) adds some aliases on top
+#   RSpec, making it more natural to do acceptance testing. It also
+#   contains some Rails generators to get tests going and organized.
+
+# Load the tools and entire application stack.
+require File.expand_path(File.dirname(__FILE__) + '/acceptance_helper')
+
+# Define the feature we are testing.
+feature "Tracking" do
+
+  # Define the scenario for the test.
+  #
+  # We need Selenium because the tracking code is written in JavaScript.
+  # The `:js => true` option will make the entire scenario run with
+  # Selenium.
+  scenario "Add the tracking code to a site", :js => true do
+    # Create a user with an email and a password.
+    Factory :user, :email => "joe@snowfinch.net", :password => "123456"
+
+    # Visit the sign in page. Path helpers are located in
+    # `spec/acceptance/support/paths.rb`.
+    visit sign_in_page
+
+    # Sign in with the previously created user.
+    fill_in "Email",    :with => "joe@snowfinch.net"
+    fill_in "Password", :with => "123456"
+    click_button "Sign in"
+
+    # Go to the site creation page and create a site.
+    click_link "Add a site"
+    fill_in "Site name", :with => "Tracking test"
+    select "Helsinki", :from => "Time zone"
+    click_button "Save"
+
+    # Grab the tracking code displayed on the page.
+    tracking_code = find("pre").text
+
+    # Gets the current URL that Selenium is accessing.
+    uri = URI.parse(current_url)
+    uri.path = "/"
+
+    # Replace all instances of the test URI with the Selenium one.
+    tracking_code.gsub!(
+      "http://snowfinch.rails.fi:3000/",
+      uri.to_s
+    )
+
+    # Create an empty HTML page that contains the tracking code.
+    html = <<-HTML
+      <!DOCTYPE html>
+      <html>
+        <head><title>Tracking test</title></head>
+        <body>#{tracking_code}</body>
+      </html>
+    HTML
+
+    # We'll store the HTML into a tempfile. Here we define its path.
+    tempfile_path = Rails.root.join("public/tracking_test.html")
+
+    # Save the page into a file that we can navigate to.
+    File.open(tempfile_path, "w") do |file|
+      file.write html
+    end
+
+    # Visit the file we just saved.
+    visit "/tracking_test.html"
+
+    # Remove the temporary HTML file.
+    FileUtils.rm(tempfile_path)
+
+    # Go back to the sites page.
+    visit sites_page
+
+    # Follow the link to the site we are tracking.
+    click_link "Tracking test"
+
+    # Check if the click got registered, by checking the counters on the
+    # site dashboard.
+    find("div[data-counter='active_visitors'] .value").should have_content("1")
+    find("div[data-counter='pageviews_today'] .value").should have_content("1")
+    find("div[data-counter='visitors_today'] .value").should have_content("1")
+  end
+end