Using jasmine with ruby
The jasmine gem can be used both with and without Rails. (Note: The jasmine gem only supports the Asset Pipeline, not Webpacker. jasmine-browser-runner works with either style of Rails JavaScript management.)
Add the jasmine gem to your gemfile and bundle install
gem 'jasmine'
Setup
To install a default jasmine.yml and a sample jasmine_helper.rb If your project is using Rails, jasmine has generators you can use to get everything set up.
rails g jasmine:install
If your project doesn’t use Rails, all of these commands also exist in the command line tool jasmine.
The command line tool will also modify your Rakefile to load the jasmine tasks
jasmine init
Jasmine also has some example specs, with implementation, that you can install.
rails g jasmine:examples
jasmine examples
Use
Once you’ve installed your jasmine.yml two commands will be available to you in rake.
If you want to start a server that continues to run so you can point a browser at it
rake jasmine
For use on your CI server
rake jasmine:ci
You can also override randomization settings from jasmine.yml for an individual jasmine:ci run for debugging purposes.
rake jasmine:ci[true]
A seed can also be specified
rake jasmine:ci[true,4321]
At this point you should be able to write your first suite
Configuration
Primary configuration is done in jasmine.yml, which is in spec/javascripts/support by default.
jasmine.yml has configuration on what files should be loaded on the page running tests as well as the location of a ruby file to be loaded for more complex configuration, demonstrated below:
The jasmine_helper.rb file specified by the spec_helper key in jasmine.yml consists of a configure block.
Jasmine.configure do |config|
# You can add rack handlers for specific urls
config.add_rack_path '/something' do
[200]
end
# And mount other rack applications
config.add_rack_app MyRackApp
# You can configure the port that the `rake jasmine` command starts a server on
config.server_port = 12345
# You can configure the port that the `rake jasmine:ci` command starts it's server on
config.ci_port = 54321
# You can add [custom formatters](#section-Custom_Formatters)
config.formatters << My::Custom::Formatter
# You can use a [custom runner](#section-Custom_Runners)
# The `runner` option on config should be a lambda or Proc that receives a formatter
# and server url and returns a constructed runner object. The lambda allows you to
# configure other options that need to be configured at initialization time.
config.runner = lambda do |formatter, server_url|
My::Custom::Runner.new(formatter, server_url, 100)
end
end
Configuring the default phantomjs runner
The phantomjs runner supports some additional options.
If you want to see the output from console.log messages from your specs, set show_console_log to true.
show_console_log: true
If you need to configure the phantomjs webpage object, you can specify a config script.
phantom_config_script: 'relative/path/from/project/root.js'
This file will be required by the phantom runner and the configure function will be passed the constructed page object.
exports.configure = function(page) {
page.viewportSize = {
width: 340,
height: 220
};
};
Custom Formatters
By default the jasmine:ci rake task outputs .s, Fs, and *s as the specs run and a summary at the end.
If you want to change the output, or produce some other output, you’ll want a custom formatter.
For an example of how to add your custom formatter see the configuration section
Your custom formatter must implement 2 methods, format and done
class My::Custom::Formatter
# `format` is called by the runner every time it gets a batch of results from the page.
# The parameter will be an array of `Jasmine::Result` objects
def format(results)
results.each do |result|
puts result.status
end
end
# `done` will be called by the runner after all results have come in.
def done
puts 'Done running tests'
end
end
The jasmine team also maintains a custom formatter that produces junit style XML for use on a CI server that knows how to parse it. Jasmine JUnit XML Formatter
Custom Runners
By default the jasmine:ci rake task uses phantomjs to load the jasmine spec runner page and run the tests.
If you want to run your tests with a different browser, or change how phantom is used, you’ll want a custom runner.
For an example of how to add your custom runner see the configuration section
Once constructed, a runner only needs to implement a run method
class My::Custom::Runner
def initialize(formatter, jasmine_server_url, result_batch_size)
# The formatter passed in is responsible for making sure all configured
# formatters receive the same messages.
@formatter = formatter
# The `jasmine_server_url` is the full http://<host>:<port> url where
# the jasmine server was started
@jasmine_server_url = jasmine_server_url
@result_batch_size = result_batch_size
end
# `run` is responsible coordinating the test run.
def run
# Here we're using Phantom to load the page and run the specs
command = "#{Phantomjs.path} 'phantom_run.js' #{@jasmine_server_url} #{@result_batch_size}"
IO.popen(command) do |output|
# The `phantom_jasmine_run.js` script writes out batches of results as JSON
output.each do |line|
raw_results = JSON.parse(line, :max_nesting => false)
# Formatters expect to get `Jasmine::Result` objects.
# It is the runner's job to convert the result objects from the page,
# and pass them to the `format` method of their formatter.
results = raw_results.map { |r| Result.new(r) }
@formatter.format(results)
end
end
# When the tests have finished, call `done` on the formatter to run any
# necessary completion logic.
@formatter.done
end
# If the runner needs some javascript to be loaded into the page as part of the load,
# it returns the full path in `boot_js`
def boot_js
File.expand_path('runner_boot.js', __FILE__)
end
end
The jasmine team also maintains a custom runner that uses selenium (and optionally SauceLabs) to run your specs with other browsers. Jasmine Selenium Runner
