Integrating with Ruby Test::Unit

Installing the Ruby Test::Unit Test Runner

This section outlines how to install the SpiraTest custom Test Runner for Test::Unit onto a workstation so that you can then run automated Test::Unit tests against a Ruby application and have the results be recorded as test runs inside SpiraTest. It assumes that you already have a working installation of SpiraTest v2.3 or later, and a working Ruby development environment. If you have an earlier version of SpiraTest you will need to upgrade to at least v2.3 before trying to use this extension.

To obtain the version of the Test::Unit test runner that is compatible with your version of SpiraTest, you simply need to log-in as a project-level administrator to SpiraTest, go to the Administration home page and download the Test::Unit test runner compressed archive (.zip). This process is described in the SpiraTest Administration Guide in more detail.

The Test::Unit test runner is provided as a set of Ruby source files that can be imported into your existing unit tests to add the SpiraTest reporting functionality. Once you have downloaded the Zip archive, you simply need to uncompress it into a folder of your choice on your local system (e.g. C:\Program Files\SpiraTest\RubyTestUnitRunner)

Now to use the custom test runner within your test cases, you need to first make sure that the folder is added to the Ruby RUBYPATH (or just the system PATH). The method for doing this is dependent on the platform you're using, so please refer to the documentation on http://ruby-lang.org for details on the appropriate method for your platform. As an example, on a Windows platform, the folder would be added to the RUBYPATH by typing the following:

set RUBYPATH=%RUBYPATH%; C:\Program Files\SpiraTest\RubyTestUnitRunner

Once you have completed this step, you are now ready to begin using your Test::Unit test fixtures with SpiraTest.

Using Ruby Test::Unit with SpiraTest

The typical code structure for a Test::Unit test suite and test case coded in Ruby is as follows:

#this is a test case that tests addition operations

class TC_Adder < Test::Unit::TestCase

def setup

@adder = Adder.new(5)

end

def test_add

assert_equal(7, @adder.add(2), "Should have added correctly")

end

def test_addfail

assert_equal(7, @adder.add(3), "Test failure")

end

def teardown

@adder = nil

end

end

#this is a test suite that calls the test case

class TS_Examples

def self.suite

suite = Test::Unit::TestSuite.new

suite << TC_Adder.suite

return suite

end

end

Test::Unit::UI::Console::TestRunner.run(TS_Examples)

The Test::Unit test case is marked as a Test::Unit test case by inheriting from the Test::Unit::TestCase base class, and the individual test methods are identified by using the 'test' prefix, with special setup and teardown methods reserved for the respective purposes. When you open up the class in a Ruby Test::Unit runner or execute from the command line it loads all the test classes and executes all the methods marked with 'test...' in turn.

Each of the Assert statements is used to test the state of the application after executing some sample code that calls the functionality being tested. If the condition in the assertion is true, then execution of the test continues, if it is false, then a failure is logged and Test::Unit moves on to the next test method.

So, to use SpiraTest with Test::Unit, each of the test cases written for execution by Test::Unit needs to have a corresponding test case in SpiraTest. These can be either existing test cases that have manual test steps or they can be new test cases designed specifically for automated testing and therefore have no defined test steps. In either case, the changes that need to be made to the Test::Unit test case and test suite for SpiraTest to record the Test::Unit test run are illustrated below:

#this is a test case that tests addition operations

class TC_Adder < Test::Unit::TestCase

def setup

@adder = Adder.new(5)

end

def test_add__2

assert_equal(7, @adder.add(2), "Should have added correctly")

end

def test_addfail__3

assert_equal(7, @adder.add(3), "Test failure")

end

def teardown

@adder = nil

end

end

#this is a test suite that calls the test case

class TS_Examples

def self.suite

suite = Test::Unit::TestSuite.new

suite << TC_Adder.suite

return suite

end

end

projectId = 1

releaseId = 2

testSetId = -1

testRunner = Test::Unit::SpiraTest::TestRunner.new(TS_Examples, "http://servername/SpiraTest", "fredbloggs", "fredbloggs", projectId, releaseId, testSetId)

testRunner.start

Firstly, each of the individual test methods is appended with two underscores followed by the ID of the corresponding test case in SpiraTest. So for example test_add is now test_add__2 as it maps to test case TC00002 inside SpiraTest.

Second, at the end of the test suite, instead of just creating the standard Console Test Runner class and passing it a reference to the test suite (e.g. TS_Examples), we now create an instance of the special Test::Unit::SpiraTest::TestRunner class, passing it a reference to the test suite as well as specifying the SpiraTest connection information.

This class takes the results from the test suite being executed and uses it to generate the web-service messages that are sent to SpiraTest to communicate the test results.

The following parameters need to be passed during the instantiation of the Test::Unit::SpiraTest::TestRunner object so that the custom test runner can access the SpiraTest repository:

suite -- the reference to the Test::Unit test suite that contains the test cases being executed. In our example above, this is the TS_Examples class.

baseUrl-- The base URL used to access your instance of SpiraTest (e.g. http://myserver/SpiraTest). It should include the protocol (e.g. http/https), the server-name, the port number (if not 80/443) and the virtual directory (if there is one).

userName - A valid username for the instance of SpiraTest that has access to the project specified above

password - A valid password for the user specified above

projectId - The ID of the project inside SpiraTest (this can be found on the project homepage in the "Project Overview" section)

releaseId - The ID of the SpiraTest release to associate the test run with. This can be found on the releases list page (click on the Planning > Releases tab). If you don't want to associate the test run with a specific release, just use the value -1 to indicate N/A.

testSetId - The ID of the SpiraTest test set to associate the test run with. This can be found on the test set list page (click on the Testing > Test Sets tab). If you don't want to associate the test run with a specific test set, just use the value -1 to indicate N/A.

Now all you need to do is save your code, launch Test::Unit, run the test fixtures as you would normally do (e.g. by executing the TS_Examples ruby file from the command line), and when you view the test cases in SpiraTest, you should see a Ruby Test::Unit automated test run displayed in the list of executed test runs:

Clicking on one of the Ruby Test::Unit test runs will bring up a screen that provides information regarding what Ruby Test::Unit test method failed, what the error was, together with the associated code stack-trace:

Congratulations... You are now able to run Test::Unit automated tests and have the results be recorded within SpiraTest. The sample test suite ts_examples.rb together with two test cases (tc_adder and tc_subtracter) is provided with the installation.