Python automation testing basics

##UI testing with Python and Selenium Webdriver Understanding the basics with data driven UI testing with Python, Selenium & nose / unittest.

Note: 'nose' is used only for standalone execution. As I use PyCharm with 'pydev' support for debugging and 'unittest' module for running the tests, I am actually not using it.

For DDT (Data driven testing) you have to install 'ddt', 'xlrd' or use csv

Automated tests framework - unittest

A test case is created by subclassing unittest.TestCase. The individual tests are defined with methods whose names start with the letters test. This naming convention informs the test runner about which methods represent tests. (see ui_test_basics.py) The setUp() and tearDown() methods allow you to define instructions that will be executed before and after each test method (used in ui_test_basics5.py).

Other Automated tests

Files ui_test_basics2.py, ui_test_basics3.py and ui_test_basics4.py are basic .py scripts demonstrating the ability to build the automated test framework with Selenium Webdriver and Python. All except ui_test_basics4.py (Safari) and ui_test_basics5 (Chrome) are using Firefox browser. As Selenium Webdriver is native written for use with Firefox, so for other browser support there is need to install additional drivers (as described below).

##Using Safari browser When you simply try to run a test using webdriver.Safari() you will get an SELENIUM_SERVER_JAR message. I found several solutions for running Selenium Webdriver with Python bindings on Safari browser but none of them worked for me. It should be easy. First, you need valid certificate from Apple. And yes, that's a major stop for me. Apple merged his Safari Developer Program with other one(s) and created Apple Developer Program which costs 99 USD per membership year. So a showstopper for me. If you have it, continue. Second, you need to download 'Selenium Standalone Server' and put it somewhere (I put it in the same directory as my python script). Then you are able to set the SELENIUM_SERVER_JAR variable: os.environ["SELENIUM_SERVER_JAR"] = "selenium-server-standalone-2.46.0.jar"
Don't forget to import os.
Third, you need to install an extension into Safari. You can use downloaded .JAR file, copy it to a directory of your choice, run jar xf selenium-server-standalone-2.46.0.jar (the filename can be different in your case) and navigate to org/openqa/selenium/safari where SafariDriver.safariextz is located. Just install & activate it (Open Safari -> Preferences -> Extensions (tab)).

Note: if you have a solution for Safari 8 on Yosemite (10.10.X) which is working, pls share it with me.
Note #2: there is a brew formula selenium-server-standalone which can be installed instead of downloading .JAR file.

##Using Chrome Before you run a python file ui_test_basics5.py it is good to read 'Chromedriver - Getting Started (https://sites.google.com/a/chromium.org/chromedriver/getting-started)' to find out how to configure the script to get chromedriver executable to work (best choice for MAC OS X is to download it to /usr/bin) and setup the PATH so you don't get any kind of error).

##Page object pattern script In PO_template_script directory you can find a template script using page object pattern.

test_case_script.py
 +-- page.py
  +-- element.py
  +-- locators.py

This approach is well-known as Page Objects Model (POM) as well. For further information and explanation how and why, please read carefully: Page Objects in Python

Screenshot functionality

In ui_test_basics5.py you will find the usage of "take screenshot on error" in the test_2_search_by_name test case. Another solution used in tearDown() method:

import sys, unittest
from datetime import datetime

class TestCase(unittest.TestCase):

    def setUp(self):
        some_code

    def test_case(self):
        test case comes here

    def tearDown(self):
        if sys.exc_info()[0]:  # Returns the info of exception being handled
            fail_url = self.driver.current_url
            print fail_url
            now = datetime.now().strftime('%Y-%m-%d_%H-%M-%S-%f')
            self.driver.get_screenshot_as_file('/path/to/file/%s.png' % now) # my tests work in parallel, so I need uniqe file names
            fail_screenshot_url = 'http://debugtool/screenshots/%s.png' % now
            print fail_screenshot_url
        self.driver.quit()

HTML report

HTMLTestRunner is an extension to the Python standard library's unittest module. It generates easy to use HTML test reports. See test_HTMLTestRunner.py for test and demo of HTMLTestRunner.py More about HTMLTestRunner: http://tungwaiyip.info/software/HTMLTestRunner.html

Data driven testing

Example of xlrd impementation - DDT/xlrd_example.py

Coming soon (maybe):

• cross-browser script
• data driven script with csv source file

Credits:

• http://selenium-python.readthedocs.org/en/latest/index.html
• http://engineering.aweber.com/getting-started-with-ui-automated-tests-using-selenium-python/
• https://code.google.com/p/selenium/wiki/PythonBindings
• http://scrolltest.com/python/selenium-testcase-with-nose-in-python/
• http://scrolltest.com/python/selenium-test-case-using-data-driven-testing-in-python/
• http://seleniummaster.com/sitecontent/index.php/selenium-web-driver-menu/selenium-test-automation-with-python-menu/188-python-selenium-web-driver-data-driven-framework
• https://docs.python.org/2/library/unittest.html
• https://code.google.com/p/selenium/wiki/SafariDriver
• https://blog.codecentric.de/en/2015/02/selenium-webdriver-safari-8/
• http://damien.co/resources/how-to-get-started-selenium-2-mac-os-x-java-7403
• https://sites.google.com/a/chromium.org/chromedriver/home
• https://pragprog.com/magazines/2010-08/page-objects-in-python