EuroPython 2009: Acceptance Testing with RobotFramework - mrtopf.demrtopf.de

EuroPython 2009: Acceptance Testing with RobotFramework

EuroPython 2009 has officially started today and the first talk I am attending is about testing, more in detail it’s about acceptance testing with RobotFramework, presented by Pekka Klärck.

RobotFramework seems to be quite interesting in that it allows you to write test cases in a very high level language which then is compiled into actual running tests. An example of a high level test case description is this:

Test Case Action Argument Argument
Valid Login Open Login Page
Input Name demo
Input Password mode
Submit Credentials
Welcome Page Should Be Open
Setting Variables Do Something first argument second argument
${value} = Get Some Value
Should Be Equal ${value} Expected value

This would be an example of a workflow based tests. Data driven tests are also possible.

Now here we have a term „Open Login Page“ which needs to be first converted into a proper action. „Open Login Page“ would be implemented like this (called user keywords):

Keyword Action Argument Argument
Open Login Page Open Browser http://host/login.html
Title Should Be Login Page
Title Should Start With [Arguments] ${expected}
${title} = Get Title
Should Start With ${title} ${expected}

Note that keywords are not single words but the whole phrase, e.g. „Open Login Page“.

The programmer would then create a library for performing the actual tests.

For instance it could look like this:

import os
import sys


class LoginLibrary:

    def __init__(self):
        self._sut_path = os.path.join(os.path.dirname(__file__),
                                      '..', 'sut', 'login.py')
        self._status = ''

    def create_user(self, username, password):
        self._run_command('create', username, password)

    def change_password(self, username, old_pwd, new_pwd):
        self._run_command('change-password', username, old_pwd, new_pwd)

    def attempt_to_login_with_credentials(self, username, password):
        self._run_command('login', username, password)

    def status_should_be(self, expected_status):
        if expected_status != self._status:
            raise AssertionError("Expected status to be '%s' but was '%s'"
                                  % (expected_status, self._status))

    def _run_command(self, command, *args):
        command = '"%s" %s %s' % (self._sut_path, command, ' '.join(args))
        process = os.popen(command)
        self._status = process.read().strip()
        process.close()

These libraries would then actually perform the tests.

You can do a lot more and there seem to be lots of way on how you can group your tests, there are many tools for reporting and you can implement those test libraries in Java or Python.

Technorati Tags: , , , , ,

Kommentare sind geschlossen.