Uses Ruby and the Ruby gems: Capybara, RSpec, Cucumber and selenium-webdriver to create a simple integration/acceptance test system for WISE4.
The tests expect an instance of WISE4 to be running at http://localhost:8080
Install rvm and then use rvm to install and use ruby 1.9.2.
rvm install 1.9.2
rvm use 1.9.2
Update the installed version of RubyGems that comes with Ruby v1.9.2. On my system this updated RubyGems from v1.6.2 to 1.8.9.
gem update --system
Install a few globally useful gems for Ruby 1.9.2
rvm gemset use global
gem install bundler rake ruby-debug19
Clone the wise4-test repository:
git clone git://github.com/concord-consortium/wise4-test.git
cd wise4-test
Create an .rvmrc file to use ruby 1.9.2 and a wise4-test gemset whenever you change to this wise4-test directory:
echo 'rvm use 1.9.2@wise4-test --create' > .rvmrc
Change out of and back into the directory to enable the .rvmrc file:
cd ..; cd -
Install the required Ruby gems into the wise4-test gemset:
bundle install
Install executable stubs into a ./bin dir so you don't always have to start a command with: bundle exec:
bundle install --binstubs
If you are using these tests scripts with a WISE4 instance running in a Vagrant/VirtualBox instance it can be helpful to be able to run remote commands on the VM instance using ssh. For example a command to reset the databases to the new default state and restart tomcat.
You need two specific pieces of data.
The ip address the vagrant/virtual-box instance is running on.
You can find this in the Vagrantfile itself. Here's an example:
config.vm.network "33.33.33.10"
The path to the vagrant ssh identity file.
This is located at the path: files/vagrant in the directory for the installed vagrant gem.
If the vagrant gem is accessible from the RVM/Ruby gemset used by wise4-test then you can display the full path to the vagarnt identity file with this command:
ruby -e "puts File.join(Gem::Specification.find_by_name('vagrant').gem_dir, 'files', 'vagrant')"
=> /Users/stephen/.rvm/gems/ruby-1.9.2-p290/gems/vagrant-0.8.2/files/vagrant
With these two items of information you can use ssh to run scripts on a running vagrant/virtual-box instance.
For example this command runs the command: /opt/ruby/bin/ruby reset_wise4_databases.rb on my wise4-vagrant instance in the /home/vagrant directory and displays the results in the console.
ssh -l vagrant 33.33.33.10 -i /Users/stephen/.rvm/gems/ruby-1.9.2-p290/gems/vagrant-0.8.2/keys/vagrant '/opt/ruby/bin/ruby reset_wise4_databases.rb'
Make sure a local wise4 server is running at http://localhost:8080. See wise4-vagrant for one easy solution.
Run all the RSpec spec tests:
rake spec
The RSpec spec tests are in the spec/ directory.
Here's an example that uses the Capaybara DSL for interacting with WISE4 via selenium-webdriver:
describe "An admin user can" do
def login(username, password)
visit "/webapp/index.html"
fill_in 'Username', :with => username
fill_in 'Password', :with => password
find("#signInButton").click
end
it "login" do
login('admin', 'pass')
find("h2").text.should eql("Welcome to the WISE Administrator Page")
end
end
Running just this tests in the console with rake produces the following results on my system:
$ rake
/Users/stephen/.rvm/rubies/ruby-1.9.2-p290/bin/ruby -S bundle exec rspec -fp --color spec/admin_login_spec.rb
.
Finished in 6.91 seconds
1 example, 0 failures
You can run a single file of spec tests like this:
rspec spec/admin_spec.rb
Run all the Cucumber feature tests:
rake features
The Cucumber features tests are in the features/ directory.
Capybara has an extensive DSL for interrogating web pages: Capybara DSL
Capybara uses css selectors by default bit also has a full set of xpath selectors. Capybara: XPath, CSS and selectors
Capybara Actions all return objects
attach_file(locator, path)Find a file field on the page and attach a file given its path.check(locator)Find a check box and mark it as checked.choose(locator)Find a radio button and mark it as checked.click_button(locator)Finds a button by id, text or value and clicks it.click_link(locator)Finds a link by id or text and clicks it.click_link_or_button(locator) (also: #click_on)Finds a button or link by id, text or value and clicks it.fill_in(locator, options = {})Locate a text field or text area and fill it in with the given text The field can be found via its name, id or label text.select(value, options = {})Find a select box on the page and select a particular option from it.uncheck(locator)Find a check box and mark uncheck it.unselect(value, options = {})Find a select box on the page and unselect a particular option from it.
Capybara Matchers all return booleans:
has_button?(locator)Checks if the page or current node has a button with the given text, value or id.has_checked_field?(locator)Checks if the page or current node has a radio button or checkbox with the given label, value or id, that is currently checked.has_content?(content)Checks if the page or current node has the given text content, ignoring any HTML tags and normalizing whitespace.has_css?(path, options = {})Checks if a given CSS selector is on the page or current node.has_field?(locator, options = {})Checks if the page or current node has a form field with the given label, name or id.has_link?(locator, options = {})Checks if the page or current node has a link with the given text or id.has_no_button?(locator)Checks if the page or current node has no button with the given text, value or id.has_no_checked_field?(locator)Checks if the page or current node has no radio button or checkbox with the given label, value or id, that is currently checked.has_no_content?(content)Checks if the page or current node does not have the given text content, ignoring any HTML tags and normalizing whitespace.has_no_css?(path, options = {})Checks if a given CSS selector is not on the page or current node.has_no_field?(locator, options = {})Checks if the page or current node has no form field with the given label, name or id.has_no_link?(locator, options = {})Checks if the page or current node has no link with the given text or id.has_no_select?(locator, options = {})Checks if the page or current node has no select field with the given label, name or id.has_no_selector?(*args)Checks if a given selector is not on the page or current node.has_no_table?(locator, options = {})Checks if the page or current node has no table with the given id or caption.has_no_unchecked_field?(locator)Checks if the page or current node has no radio button or checkbox with the given label, value or id, that is currently unchecked.has_no_xpath?(path, options = {})Checks if a given XPath expression is not on the page or current node.has_select?(locator, options = {})Checks if the page or current node has a select field with the given label, name or id.has_selector?(*args)Checks if a given selector is on the page or current node.has_table?(locator, options = {})Checks if the page or current node has a table with the given id or caption.has_unchecked_field?(locator)Checks if the page or current node has a radio button or checkbox with the given label, value or id, that is currently unchecked.has_xpath?(path, options = {})Checks if a given XPath expression is on the page or current node.
The following Capybara Finders all return Capybara::Element objects:
all(*args)Find all elements on the page matching the given selector and options.find(*args)Find an Element based on the given arguments.find_button(locator)Find a button on the page.find_by_id(id)Find a element on the page, given its id.find_field(locator) (also: #field_labeled)Find a form field on the page.find_link(locator)Find a link on the page.
The first Capybara finder returns an object:
first(*args)Find the first element on the page matching the given selector and options, or nil if no element matches.
If you want to debug a test with the console-based Ruby debugger insert the following line in your spec test:
debugger
And start the spec test with debugging turned on with the -d option:
rspec spec/admin_spec.rb -d
Your test will execute until the debugger statement at which point the Ruby debugger will be open in the console and FireFox will be on the screen.
When you are finished debugging complete the test by entering: cont
Insert this line into a test to save and open a copy of the html page:
save_and_open_page