DEV Community: Md Zinnun Uddin

Getting Started with Selenium Grid

Md Zinnun Uddin — Wed, 24 Aug 2022 04:46:00 +0000

I was learning docker to containerize automation testcases. Moving forward, I realized- I don't even have the basics and it is difficult to setup and run testcases without knowing the fundamentals of Selenium Grid.

So, I decided to start learning grid first, then coming back to docker. Reading this blog you will know about grid from very basic to some basics 😄

Step 01: Configure

Assuming-

Java is already installed.
Environment variable has been set.

To start handling with grid, we need selenium server standalone JAR file. We are going to use selenium-server-standalone-3.141.59. Download it from here.

Here comes Hub and Node concept. A Hub is the central point where nodes will be connected. Hub is something like a docking system, where we can connect several devices via different ports/interfaces to a system. In selenium grid- a node may contain several browsers.

First, make a directory and put the selenium-server-standalone-3.141.59.jar there. This step is very important, because every time we want to start a hub- we must keep the jar file in the same directory.

Step 02: Start Hub

Open a cmd terminal window and type-

java -jar selenium-server-standalone-3.141.59.jar -role hub

💡 Note: By default, it connects to port 4444. We can use any port using -port parameter.

java -jar selenium-server-standalone-3.141.59.jar -role hub -port 5555

now, see the grid console at http://localhost:4444/grid/console/

The console is blank as of now, because no node is registered yet.

Step 03: Register Node

Now, we need to register some nodes to the hub. Open a new cmd window(in the same directory) and type-

java -jar selenium-server-standalone-3.141.59.jar -role node -hub http://localhost:4444/grid/register

A new message apperared in hub cmd after executing the command, see below-

💡 Note: We must put chromedriver.exe/ geckodriver.exe/ iedriver.exe in the same directory.

Registering a node has been succeeded, refresh the window in browser-

By default, it creates 11 browser instances (5 chrome, 5 firefox & 1 internet explorer) and it connects with hub using port 4444. To register to a particular port, you can use -port parameter as said before.

localhost and IP address:

We are setting up hub and nodes on the same local machine. So that, localhost has been used instead of machine IP address. But in real situation, nodes will be connected from different machines with the hub. To find the IP of your machine-

open cmd
type ipconfig & press Enter

the highlighted IPv4 Address is our machine ip-

localhost will be replaced with IP address along with port no. to register node from a remote machine like this-

java -jar selenium-server-standalone-3.141.59.jar -role node -hub http://192.168.251.61:4444/grid/register

To create only one instance of a particular browser-

java -jar selenium-server-standalone-3.141.59.jar -role node -hub http://localhost:4444/grid/register -browser browserName=chrome

We can create any browser instance using -browser browserName=chrome parameter. Change any browser name instead of chrome.

To create single instance of several browser-

java -jar selenium-server-standalone-3.141.59.jar -role node -hub http://localhost:4444/grid/register -browser browserName=chrome -browser browserName=safari -browser browserName=firefox

Refresh the grid console, the console should look like this-

This is the end of the first tutorial on Selenium Grid. In the next tutorial, we will run a dummy project and learn about- how to use parameters and capabilities to define browser, browser version, max instances, max sessions etc.

Thanks for reading!

Page Object Model & PageFactory in Selenium

Md Zinnun Uddin — Mon, 11 Jul 2022 08:29:00 +0000

✒️ What is Page Object Model?

POM is a design pattern in Selenium that creates an Object Repository to store web elements. In POM, we follow the principle of separation of Object Repository and Test Classes. Each object repository only contains the elements of the respective webpage.

✒️ What are the advantages of POM?

✔️ Easy Maintenance: As every page is separate in POM, if there is any change in any webpage that is easy to identify and change. Changes in the object repository don't impact the code elsewhere.

✔️ Reusability: A single object repository can be used for different TestCases. Also, we can integrate the same POM in Selenium with different types of tests like functional testing and acceptance testing simultaneously.

✔️ Readability: Testers can easily identify actions that will be performed on a particular webpage.

⚠️ verifications or assertions shouldn't belong to Object Repository ⚠️

✒️ What is PageFactory in Selenium?

PageFactory is a very optimized way of implementing POM. It is a class provided by Selenium WebDriver which expands the functionality of plain POM.

1️⃣ @FindBy: An annotation used in Page Factory to declare web elements. The @FindBy annotation supports :id, name, className, css, tagName, linkText, partialLinkText, xpath. Below is an example of declaring an webelement using @FindBy-

@FindBy(class="elementClass")
WebElement element;

2️⃣ initElements(): A static method in Page Factory class, it instantiate an instance of the given class. Also used to initialize all the web elements defined by @FindBy annotation.

public RegistrationPage(WebDriver driver) {           
         this.driver = driver; 
         PageFactory.initElements(driver, this);
}

3️⃣ AjaxElementLocatorFactory: AjaxElementLocatorFactory is Lazy Loading concept in PageFactory. It is used to deal with dynamic elements that change during runtime based on user actions, especially on Ajax-heavy applications. Timeout is also possible to set for a webelement using AjaxElementLocatorFactory.

AjaxElementLocatorFactory ajaxFactory = new AjaxElementLocatorFactory(driver, 30);
PageFactory.initElements(ajaxFactory, this);

✒️ Default behavior of PageFactory:

Web elements declared using @FindBy are initialized at once and yet not searched in the browser DOM
WebDriver searches a particular webelement each and every time it is called or any action happened to it.
For static elements, it provides @CacheLookup annotation to cache web elements for future uses.

✒️ Differences between POM and PageFactory:

Page Object Model	PageFactory
Page Object Model is a design pattern.	PageFactory is a `class` that provides an implementation of POM .
In POM, individual initialization of every page object is needed.	In PageFactory, all the page objects are initialized at once by using the initElements() method.
Define web elements using `By`	Define web elements using `@FindBy` annotation.
POM is not optimal as it doesn't support lazy initialization.	Supports lazy initialization and it is optimal.
POM is not capable of handling `StaleElementReferenceException`.	It effeciently handles `StaleElementReferenceException` by searching everytime when any stale element is called.

✒️ @CacheLookUp annotation:

@CacheLookUp annotation is used to keep a cache of WebElement on its first arrival on the webpage instead of searching over and over again, usually used for elements that never change.

A findElement() REST Request is sent to the WebDriver every time a WebElement is called from page object repository. This satisfies one of the important Pagefactory benefits of looking for the newest version of webelements after page loading. But, it makes the process slow and time-consuming(a lot obviously!). Here, @CacheLookUp comes to solve the problem.

✒️ When not to use @CacheLookUp?

It is not suitable for elements that are dynamic in nature. For AJAX based applications, it may not work where the DOM frequently changes based on user interaction. Upon using @CacheLookUp annotation, WebDriver may throw StaleElementExceptions.