Get Even More Visitors To Your Blog, Upgrade To A Business Listing >>

Automate Amazon: Writing a Sign In Test

This post is fourth of a series of nine. Need to go back to the beginning?

Drafting a Login Test for won't be as easy as drafting one for Dave Haeffner's mock site, The-Internet. There needs to be a lot more infrastructure put in place besides the CommonUtils library we worked on in the last blog post. Also, use the word "Login". Instead, they use the phrase Sign In.

Sketch the SignIn Steps

Here's how you log into sign into's site:
  1. Go to's home page,
  2. Hover your cursor over the "Your Account" block, and click the sign in button.
  3. Once you are on the sign in page, enter the username, password into the appropriate text boxes. 
  4. Select the Sign In button. 
I noticed after following this process a few times, that there sometimes was a bit of a lag, so we might want to wait until the Username and password actually appear before entering a username and password. 

Review: How to find selectors of Web Elements

Let's say we want to know about what selectors to use for the username, password, and Sign In button on the Sign In page.

You can find the ids and cssSelectors of an object by:
  • Opening the Firefox browser with the Firebug and Firepath plugin. 
  • Going to and pressing the "Sign In" button.
  • Right clicking and selecting on the textboxes or the button and selecting "Inspect in Firepath".
  • With the "Firepath" tab, make sure "CSS" is selected. 
We can see that the following web elements have these values:
  • Email textbox: cssSelector #ap_email
  • Password textbox: cssSelector #ap_password
  • Sign In Button: cssSelector #signInSubmit
... Yes, these values are technically IDs and not CSS Selectors. Adding the '#' makes them CSS Selectors. At my workplace, I find that IDs of web elements are constantly changing, since the ID is usually a numbered index. CSS Values seem to change less, so I got in the habit of leaning towards CSS Selectors over IDs. 

Sketch out the Infrastructure Needed

Judging from what we have listed above, we will need the following classes written:
  • A Test Class, PurchaseOrderTest, that stores the tests.
  • A class to store the username and password of test users, and a class to retrieve the data.
  • Methods in CommonUtilites that handle sending keys to textboxes, clicking on buttons, and hovering-and-clicking on sections of the home page.
  • Two classes called Page Objects to store how the HomePage and the LoginPage interacts with the DOM.
  • We can bundle together commonly used methods in an Actions class, and call it OrderActions.

What our Directory Structure Will Look Like

After we store the User Properties, create the Actions classes, and create the test class, our directory structure could look like:


  • actions
    • OrderActions
  • base
    • LoadProperties
  • pages
    • HomePage
    • SignInPage
  • properties
  • testcases
    • PurchaseOrderTest
  • utils
    • CommonUtils
    • DriverUtils

Store the User Properties

Store the Username and Password in a text file called a Properties file, Create a class called LoadProperties in order to handle the reading and the writing of the file. (Code formatting from

java / properties /
1:  [email protected]  
2: tester23.password=amzntester23

java / base /
1:  package base;  
2: import org.apache.commons.lang3.StringUtils;
3: import;
4: import;
5: import;
6: import java.util.Properties;
7: import java.util.Set;
8: /**
9: * Created by tmaher on 12/22/2015.
10: */
11: public class LoadProperties {
12: public static Properties user = loadProperties("src/test/java/properties/");
13: private static Properties loadProperties(String filePath) {
14: Properties properties = new Properties();
15: try {
16: FileInputStream f = new FileInputStream(filePath);
17: properties.load(f);
18: } catch (FileNotFoundException e) {
19: e.printStackTrace();
20: } catch (IOException e){
21: e.printStackTrace();
22: }
23: return properties;
24: }
25: public static String getPropertyValue(String path, String key){
26: Properties p = loadProperties(path);
27: String result = "";
28: Set<String> values = p.stringPropertyNames();
29: for(String value : values){
30: if(StringUtils.equalsIgnoreCase(value, key)){
31: result = p.getProperty(value);
32: break;
33: }
34: }
35: return result;
36: }
37: }
All Source Code can be found in T.J. Maher's Automate-Amazon repository. 

Next, we can work from the bottom level to the top:
  • Adding methods to CommonUtils
  • Abstracting each page into a PageObject such as HomePage and SignInPage
  • Bundling the methods in the PageObjects to create repeatable Actions we can use.
  • Compose the Actions methods into a test we can run. 

1. Add to CommonUtil 

Add to CommonUtils methods the basic building blocks our page objects will call. All Selenium WebDriver calls will be encapsulated in a method on CommonUtils.

  • Navigate to a URL

 public void navigateToURL(String URL) {  
try {
} catch (Exception e) {
System.out.println("FAILURE: URL did not load: " + URL);
throw new TestException("URL did not load");

  • SendKeys to a Textbox, given the web element by selector, and the value.

 public void sendKeys(By selector, String value) {  
WebElement element = getElement(selector);
try {
} catch (Exception e) {
throw new TestException(String.format("Error in sending [%s] to the following element: [%s]", value, selector.toString()));

  • Click on a button, given the web element by selector

 public void click(By selector) {  
WebElement element = getElement(selector);
try {;
} catch (Exception e) {
throw new TestException(String.format("The following element is not clickable: [%s]", selector));

  • Hovers a cursor over a web element and click it
 public void scrollToThenClick(By selector) {  
WebElement element = _driver.findElement(selector);
actions = new Actions(_driver);
try {
((JavascriptExecutor) _driver).executeScript("arguments[0].scrollIntoView(true);", element);
} catch (Exception e) {
throw new TestException(String.format("The following element is not clickable: [%s]", element.toString()));

  • Waits Until an Element is Visible
 public void waitForElementToBeVisible(By selector) {  
try {
wait = new WebDriverWait(_driver, timeout);
} catch (Exception e) {
throw new NoSuchElementException(String.format("The following element was not visible: %s", selector));

2. Page Objects

Now that the Selenium WebDriver functionality has been encapsulated, we can extend the page objects to use CommonUtils. We are inheriting from CommonUtilites the above methods.

These Page Objects are the only places that should interact with the web elements on the page, evaluating if the web elements are there, checking and setting their values.

On the top of each Page Object, we have the selectors for each web element. They are each declared private because they are only to be accessed in this page object. They are also declared final because they are constant.

By is a reserved word in the Selenium library. Web elements can be found, By.cssSelector,, etc.

1:  package pages;  
2: import enums.Url;
3: import org.openqa.selenium.By;
4: import utils.CommonUtils;
5: /**
6: * Created by tmaher on 12/21/2015.
7: */
8: public class HomePage extends CommonUtils {
9: private final By YOUR_ACCOUNT ="nav-link-yourAccount");
10: private final By SHOPPING_CART_ICON = By.cssSelector("#nav-cart");
11: private final By SHOPPING_CART_COUNT = By.cssSelector("#nav-cart > #nav-cart-count");
12: public HomePage(){
13: }
14: public void navigateToHomePage() {
15: String url = Url.BASEURL.getURL();
16: System.out.println("Navigating to " + url);
17: navigateToURL(url);
18: }
19: public void navigateToSignInPage(){
20: System.out.println("HOME_PAGE: Selecting [YOUR_ACCOUNT] in navigation bar.");
21: scrollToThenClick(YOUR_ACCOUNT);
22: System.out.println("HOME_PAGE: Navigating to the SIGNIN_PAGE.\n");
23: }
24: }

1:  package pages;  
2: import org.openqa.selenium.By;
3: import utils.CommonUtils;
4: /**
5: * Created by tmaher on 12/21/2015.
6: */
7: public class SignInPage extends CommonUtils {
8: private final By USERNAME = By.cssSelector("#ap_email");
9: private final By PASSWORD = By.cssSelector("#ap_password");
10: private final By SIGNIN_BUTTON = By.cssSelector("#signInSubmit");
11: public void enterUsername(String userName){
12: System.out.println("SIGNIN_PAGE: Entering username: " + userName);
13: waitForElementToBeVisible(USERNAME);
14: sendKeys(USERNAME, userName);
15: }
16: public void enterPassword(String password){
17: System.out.println("SIGNIN_PAGE: Entering password.");
18: waitForElementToBeVisible(PASSWORD);
19: sendKeys(PASSWORD, password);
20: }
21: public void clickSignInButton(){
22: System.out.println("SIGNIN_PAGE: Clicking the [SIGN_IN] button.\n");
23: click(SIGNIN_BUTTON);
24: }
25: }

3. Actions Classes

When someone enters a username into the SignIn page, it's a safe bet that they are going to want to enter a password, too, along with clicking on the SignIn button. If someone wants to write a test that explicitly asserts that yes, we can enter text into the username textbox, they can do that by accessing the SignIn page object. But most of the time, those three methods will be grouped together.

Below, we group them together in a method called loginAs(String username, String password). Pass in a username and password, and it will do the rest.
1:  package actions;  
2: import pages.*;
3: /**
4: * Created by tmaher on 12/21/2015.
5: */
6: public class OrderActions {
7: public void navigateToHomePage(){
8: HomePage homePage = new HomePage(); // Instantiate a new HomePage page object
9: homePage.navigateToHomePage();
10: }
11: public void loginAs(String username, String password){
12: HomePage homePage = new HomePage(); // Instantiate a new HomePage page object
13: SignInPage signIn = new SignInPage(); // Instantiate a new SignInPage page object
14: homePage.navigateToSignInPage();
15: signIn.enterUsername(username); // Use the methods on the SignInPage
16: signIn.enterPassword(password);
17: signIn.clickSignInButton();
18: }
19: }

4. PurchaseOrderTest: test_Login

Now that we have all the components, we can start assembling the pieces in the Actions class to form a test.
1:  package testcases;  
2: import base.LoadProperties;
3: import org.openqa.selenium.WebDriver;
4: import actions.*;
5: import org.testng.annotations.AfterClass;
6: import org.testng.annotations.BeforeClass;
7: import org.testng.annotations.Test;
8: import utils.DriverUtils;
9: import static org.testng.Assert.assertEquals;
10: /**
11: * Created by tmaher on 12/14/2015.
12: */
13: public class PurchaseOrderTest {
14: public WebDriver driver;
15: @BeforeClass
16: public void setUp(){
17: driver = DriverUtils.getDriver();
18: }
19: @Test()
20: public void test_Login(){
21: OrderActions orderActions = new OrderActions();
22: String username = LoadProperties.user.getProperty("tester23.username");
23: String password = LoadProperties.user.getProperty("tester23.password");
24: orderActions.navigateToHomePage();
25: orderActions.loginAs(username, password);
27: }
28: @AfterClass
29: public void tearDown(){
30: driver.quit();
31: }
32: }
All Source Code can be found in T.J. Maher's Automate-Amazon repository. 

5. Run the Test

Now that everything is all set, we can right click on the test method for test_Login and run the test:

 Navigating to  
HOME_PAGE: Selecting [YOUR_ACCOUNT] in navigation bar.
HOME_PAGE: Navigating to the SIGNIN_PAGE.
SIGNIN_PAGE: Entering username: [email protected]
SIGNIN_PAGE: Entering password.
SIGNIN_PAGE: Clicking the [SIGN_IN] button.
All Source Code can be found in T.J. Maher's Automate-Amazon repository. 

... Normally, we would then assert if we are on the correct page or not. I didn't add that into the test since what I really wanted to do was the next blog entry.

Until then, happy coding!

View the (mostly) Completed Test Code:

  • All test code can be found on GitHub at Automate-Amazon.

NEXT: Setup Objects to Handle Various Products, such as Books! >> 

-T.J. Maher
 Sr. QA Engineer, Fitbit
 Boston, MA

// Automated tester for [ 8 ] month and counting!

Please note: 'Adventures in Automation' is a personal blog about automated testing. It is not an official blog of 

Automate Amazon:
  • Introduction
  • Part One: Environment Setup
  • Part Two: Sketch Use Case
  • Part Three: CommonUtils, methods, exceptions
  • Part Four: Write Sign In Test
  • Part Five: Product Enums and Objects
  • Part Six: Initializing Login and Cart
  • Part Seven: Writing Shopping Cart Test
  • Part Eight: Data Driven Tests with TestNG XML
  • Part Nine: Code Review Request, please! 
  • Source Code: GitHub, T.J. Maher

This post first appeared on Adventures In Automation, please read the originial post: here

Share the post

Automate Amazon: Writing a Sign In Test


Subscribe to Adventures In Automation

Get updates delivered right to your inbox!

Thank you for your subscription