{"id":93927,"date":"2019-07-10T10:00:47","date_gmt":"2019-07-10T07:00:47","guid":{"rendered":"http:\/\/www.javacodegeeks.com\/?p=93927"},"modified":"2019-07-03T16:54:36","modified_gmt":"2019-07-03T13:54:36","slug":"test-automation-selenium-webdriver","status":"publish","type":"post","link":"https:\/\/www.javacodegeeks.com\/2019\/07\/test-automation-selenium-webdriver.html","title":{"rendered":"22 Practical Tips To Test Automation With Selenium WebDriver"},"content":{"rendered":"

Test automation with Selenium has empowered website testers over the globe to perform automated website testing with ease. Webdriver is a core component of the Selenium framework using which you can perform automated cross browser testing <\/a>of your website or web application against different types of browsers e.g. Google Chrome, Mozilla Firefox, Safari, Opera, Internet Explorer, Microsoft Edge, etc.<\/p>\n

The primary advantage of performing test automation with Selenium Webdriver over other web automation tools\/frameworks is the support for a wide number of programming languages namely Python, Java, C#, Ruby, PHP, JavaScript, .Net, Perl, etc. If you are new to automation testing with Selenium WebDriver, then you can have a look at our Selenium WebDriver tutorial for automated cross browser testing<\/a> where we talk about the overall architecture of Selenium and how the framework can be used with popular programming languages. You can also check my previous article on Selenium Grid setup Tutorial for cross browser testing<\/a>, to leverage the ability of parallel testing with Selenium<\/a>. Irrespective of the programming language being used, there are certain best practices that are applicable for performing test automation with Selenium Webdriver (independent of the development language).<\/p>\n

In this article, I will share with you some key tips for Selenium automation testing, that touch upon aspects of code optimization, performance improvements, dynamic web-page loading, handling CSS and HTML code, etc.<\/p>\n

Note<\/b> \u2013 Majority of these coding tips for test automation with Selenium WebDriver are generic in nature and can be applied irrespective of the programming language used for the development of test scripts. However, for below demonstration, we have made use of Selenium with Python language.<\/p>\n

Selenium Tip #1 \u2013 Setting The Executable Path Of The Selenium Webdriver<\/h2>\n

In order to communicate with the browser under test, you need to first download the corresponding plugin\/webdriver from their official website<\/a>. This plugin would be responsible for communicating with the browser and it should be present in your machine (on which you are developing the tests). The plugin\/webdriver path has to be set in the Selenium Webdriver configuration.<\/p>\n

Though the plugin\/webdriver can be placed in any location since you can provide the static\/relative path in the Selenium Webdriver configuration, the approach can be error prone and you need to keep track of the file path. A better & more reliable approach is to place the corresponding Selenium Webdriver in the location where the driver executable is present, in which case you need not specify the executable path in the Selenium Webdriver configuration.<\/p>\n

\n
\"Selenium<\/figure>\n<\/div>\n

If the geckodriver<\/b> is not present in the Browser location, you need to manually add the path of the same in the source code. We import the selenium.webdriver.firefox.firefox_binary<\/b> module to provide the path to the Firefox executable.<\/p>\n

from selenium import webdriver\nfrom selenium.webdriver.firefox.firefox_binary import FirefoxBinary\n \nff_binary = FirefoxBinary('path\/to\/gecko driver')\nbrowser = webdriver.Firefox(firefox_binary=ff_binary)\n<\/pre>\n
As seen in the code snippet below, we have not specified the location of the geckodriver (Firefox Webdriver) since it is placed in the same location where Firefox browser is present. This is more reliable approach compared to the previous one and can help in reducing basic errors in implementation of test automation with Selenium.<\/p>\n
''' Import the required modules for development '''\nfrom selenium import webdriver\nfrom selenium.webdriver.common.keys import Keys\nfrom time import sleep\n \n'''Creation of Firefox Webdriver '''\ndriver = webdriver.Firefox()\ndriver.get(\"https:\/\/www.lambdatest.com\/\")\n<\/pre>\n
Selenium Tip #2 \u2013 Capture Screenshot Of Test Automation With Selenium WebDriver<\/h2>\n
While performing tests, you would have come across requirements where a screenshot has to be captured for verifying results of the tests. Selenium WebDriver offers three APIs through which you can take a screenshot of a web-page.<\/p>\n
    \n
  1. save_screenshot(\u2018path-where-screenshot-should-be-saved\/filename.png\u2019)<\/li>\n
  2. get_screenshot_as_file(\u2018path-where-screenshot-should-be-saved\/filename.png\u2019)<\/li>\n
  3. get_screenshot_as_png()<\/li>\n<\/ol>\n
    The first two APIs lets you save the screen of the current window in a .png file. The API returns False if there is an IOError, else it returns True. These APIs would only work if the file extension is .png, else Python throws an error & the saved content might not be viewable. If you wish to capture a screen of your current window in a binary format then make use of get_screenshot_as_png() API.<\/p>\n
    ''' Import the required modules for development '''\nfrom selenium import webdriver\nimport StringIO\nfrom PIL import Image\n \n'''Creation of Firefox Webdriver '''\ndriver = webdriver.Firefox()\ndriver.get(\"https:\/\/www.lambdatest.com\/\")\n \n'''Taking screenshot of the web-page. File would be saved in the location where the source code is present '''\n \n'''Option - 1'''\ndriver.save_screenshot('screenshot_1.png');\n \n \n'''Option - 2'''\ndriver.get_screenshot_as_file('screenshot_2.png');\n'''Option - 3'''\nscreenshot = driver.get_screenshot_as_png();\n    \nscreenshot_size = (20, 10, 480, 600)\nimage = Image.open (StringIO.StringIO(screen))\nregion = image.crop(screenshot_size)\nregion.save('screenshot_3.jpg', 'JPEG', optimize=True)\n<\/pre>\n
    Selenium Tip #3 \u2013 Refreshing WebPage While Automation Testing With Selenium WebDriver<\/h2>\n
    There might be scenarios where there is a requirement for refreshing a web page, especially while waiting for a specific condition. There are a number of ways through which a webpage can be refreshed while performing test automation with Selenium Webdriver, the popular one is listed below.<\/p>\n
    1. driver.refresh() method<\/h3>\n
    As the name signifies, the refresh()<\/b> method is used to refresh the web page. It is asynchronous in nature hence; you should make use of this API in conjunction with document.readyState()<\/b>.<\/p>\n
    ''' Import the required modules for development '''\nfrom selenium import webdriver\n \n'''Creation of Firefox Webdriver '''\ndriver = webdriver.Firefox()\ndriver.get(\"https:\/\/www.lambdatest.com\/\")\ndriver.refresh()\n<\/pre>\n
    2. ActionChains() method<\/h3>\n
    ActionChains() are another way to automate low-level interactions for automation testing with Selenium, such as key press, mouse button actions, etc. In order to refresh the webpage, we make use of the \u2018CTRL + F5\u2019 combination.<\/p>\n
    import time\nfrom selenium import webdriver\nfrom selenium.webdriver.common.action_chains import ActionChains\nfrom selenium.webdriver.common.keys import Keys\n \n'''Creation of Firefox Webdriver '''\n# driver = webdriver.Chrome()\ndriver = webdriver.Firefox()\ndriver.get(\"https:\/\/www.lambdatest.com\/\")\n \ntime.sleep(5)\n \nprint(\"Before refresh\")\n \nActionChains(driver) \\\n    .key_down(Keys.CONTROL) \\\n    .send_keys(Keys.F5) \\\n    .key_up(Keys.CONTROL) \\\n    .perform()\n \nprint(\"After refresh\")\n \nsleep(5)\ndriver.quit()\n<\/pre>\n
    Selenium Tip #4 \u2013 Open A Webpage In A New Tab<\/h2>\n
    execute_script can be used to synchronously execute JavaScript code in current window\/frame. Argument (a JavaScript) to open the webpage is passed as the argument to execute_script<\/p>\n
    from selenium import webdriver\nfrom selenium.webdriver.common.keys import Keys\nfrom time import sleep\n \ndriver = webdriver.Firefox()\ndriver.get(\"http:\/\/www.google.com\/\")\n \ndriver.implicitly_wait(10)\n \n#open tab\ndriver.execute_script(\"window.open('https:\/\/www.lambdatest.com', 'new tab')\")\n \nsleep(5)\ndriver.quit()\n<\/pre>\n
    Selenium Tip #5 \u2013 Saving Partial Screenshot Of A Web page<\/h2>\n
    There are cases where you might need to take a partial screenshot of a webpage as your perform test automation with Selenium. In such cases, you can make use of the Pillow module. You need to first install the Pillow\/PIL module using the command<\/p>\n
    pip install pillow<\/code><\/p>\n
    Screenshot of the entire webpage is taken using the get_screenshot_as_png()<\/b> API. Once the screenshot is ready, the PIL library is used to open the captured image in memory, after which the image (which contains the entire webpage screenshot) is cropped to get the resultant image.<\/p>\n
    from selenium import webdriver\n''' Install the Pillow module using the command pip install pillow '''\nfrom PIL import Image\nfrom io import BytesIO\n \ndriver = webdriver.Firefox()\ndriver.get('http:\/\/google.com\/')\n \n# Use the Inspection tool to find the location of the logo\nelement = driver.find_element_by_id('hplogo')\nimage_location = element.location\nsize = element.size\n \npng = driver.get_screenshot_as_png()\n \n''' Since the webpage screenshot is ready, we can exit the browser.'''\ndriver.quit()\n \n''' PIL Library is used to open the image in memory '''\ncrop_image = Image.open(BytesIO(png))\n \n''' Extract the Left, Right, Top, and Bottom co-ordinates ''' \n \nleft = image_location['x']\ntop = image_location['y']\nright = image_location['x'] + size['width']\nbottom = image_location['y'] + size['height']\n \ncrop_image = crop_image.crop((left, top, right, bottom))\ncrop_image.save('logo-screenshot.png')\n<\/pre>\n
    Selenium Tip #6 \u2013 Execute JavaScript Code<\/h2>\n
    execute_script is used to execute JavaScript code as you perform test automation with Selenium WebDriver. The syntax is driver.execute_script(\u201cjavascript code here\u201d)<\/b>.<\/p>\n
    \n
    \"Selenium<\/figure>\n<\/div>\n
    As shown in the example below, a on_click action of Register is performed [Class name is home-cta].<\/p>\n
    from selenium import webdriver\nfrom time import sleep\n \ndriver = webdriver.Firefox()\ndriver.get(\"https:\/\/www.lambdatest.com\")\n \ndriver.execute_script(\"document.getElementsByClassName('home-cta')[0].click()\")\n \nsleep(10)\n \ndriver.close()\n<\/pre>\n
    Selenium Tip #7 \u2013 Extracting Results Of JavaScript Code<\/h2>\n
    After invoking the JavaScript code for automation testing with Selenium, you need to extract the results of these JavaScript codes. You can make use of the return keyword in order to get the result of a JavaScript code as shown in the extended example where we explained JavaScript.<\/p>\n
    from selenium import webdriver\nfrom time import sleep\n \ndriver = webdriver.Firefox()\ndriver.get(\"https:\/\/www.lambdatest.com\")\n \ndriver.execute_script(\"document.getElementsByClassName('home-cta')[0].click()\")\n    \nresult = driver.execute_script(\"return 0\")\nprint(result)\n \nsleep(10)\n \ndriver.close()\n<\/pre>\n
    Selenium Tip #8 \u2013 Handling Multiple Browser Types For Automated Cross Browser Testing<\/h2>\n
    There are multiple scenarios that you might need to test your code against different browsers e.g. Firefox, Chrome, Internet Explorer. The practice of testing a website across different browsers is termed as automated browser testing<\/a>. To perform automated browser testing with Selenium automation testing, you should incorporate selective handling of those browsers in your unittest code or pytest code. A code snippet (which makes use of pytest) to handle multiple browsers is shown below:
     
    <\/div> <\/div><\/p>\n
    # Import the 'modules' that are required for execution\n \nimport pytest\nfrom selenium import webdriver\nfrom selenium.webdriver.chrome.options import Options\nfrom selenium.webdriver.common.keys import Keys\nfrom time import sleep\n \n#Fixture for Firefox\n@pytest.fixture(params=[\"chrome\", \"firefox\"],scope=\"class\")\ndef driver_init(request):\n    if request.param == \"chrome\":\n        # Perform necessary actions here\n    if request.param == \"firefox\":\n    # Perform necessary actions here\n    yield\n    web_driver.close()\n    ...........\n    ...........\n<\/pre>\n
    \n
    \"\"<\/a><\/figure>\n<\/div>\n
    Selenium Tip #9 \u2013 Locating Elements On A Web Page Using CSS Locators<\/h2>\n
    As you perform test automation with Selenium, locating web elements on a page comes as a foundation to your automation scripts. In case you want to perform conditional execution based on the presence of a particular kind of web element like Tag, Class, ID, etc. you can make use of find_elements_*** API. Some of them are mentioned below<\/p>\n