Canary aims to provide a flexible framework for writing systems tests for web applications.
- QA, Analysts, Devs, and Management can easily understand the test results
- Devs can get a log of context test context for easy debugging
- Scripting Devs can leverage reusable business behavior with ease
Canary requires a special fork of Poltergeist which is not hosted as a rubygem. This complicates the setup process. First git clone Canary and create an environment variable CANARY_PATH = <canary_repo_path>/lib
Modify your script_helper.rb to include the following lines of code.
$:.push File.expand_path(ENV['CANARY_PATH'])
require 'canary'Canary uses two a yaml configuration files during initialization. You can configure the path where the configuration files are located from your script_helper.rb with the following.
Canary.config_path = File.expand_path(File.dirname(__FILE__))The config.yaml file can contain multiple variations of config settings. The configuration set used is determined by the following order of precedence.
ENV['COMPUTERNAME']ENV['USERNAME']'local'
Example configuration files can be found in the examples directory.
Scripts can interact with pages via making calls to page objects through the Story object. The script.rb file includes helper methods for use in your specs. The Capybara Node object is made accessible by using the method find(element_name_or_selector), element(element_name), or elements(collection_name). All method calls to the object returned by these methods are logged and made available in the test reports.
Database queries should be a subclass of Canary::Data::SQLData.
- Tests are currently written using RSpec.
- Web automation uses Capybara.
- Supports PhantomJS with Poltergeist.
- Remote command execution uses Powershell v2 Invoke-Command.
- Data access performed using MS SQL.
- Test results stored in MongoDB. Requires independent installation.
- Reports hosted using NodeJS web server. Requires independent installation.