As you start using Selenium, you will realize that there are a wide variety of things that you can do with it. In this section, you will learn about the 5 most important Selenium IDE and WebDriver programming features that you will certainly want to know about. Here, you will learn how to work with running tests on different types of web browsers, locating elements, working with WebElements, synchronization, and capturing screenshots during test runs.
One of the key features of Selenium is support for running tests on a variety of web browsers available. You can develop a test on one browser and run it on all the supported browsers as needed. This feature is essential for cross-browser testing of your web application. You can run tests on combinations of operating systems and web browsers.
Option 1 – running Selenium IDE tests on various browsers
In the Quick start section, we created a test in Selenium IDE; however, as we saw, it only allows us to record and reply tests in Firefox. We can use the Selenium standalone server to run tests that are created with Selenium IDE on different browsers other than Firefox using the following steps:
Download the latest release of Selenium standalone server from http://code.google.com/p/selenium/downloads/list.
Open a command prompt/console window and type the following command:
java -jar selenium-server-standalone-2.29.0.jar -htmlSuite *googlechrome http://demo.magentocommerce.com/ C:\SearchTests.html C:\Result.html
Note
We have used
*googlechrome
in the preceding command to run the tests in Google Chrome. You can also use*iexplore
(Internet Explorer),*safari
(Safari), and*firefox
(Firefox) to run tests on these browsers.This command will launch the Selenium standalone server. We used the
–htmlSuite
option through which we can tell the server to execute the tests created using Selenium IDE. We also specified the browser on which we want to run the tests and the name of the test suite along with the path and name for the output result file. The server will launch Google Chrome and execute all the tests from the specified test suite.
Option 2 – running Selenium WebDriver tests on various browsers
Selenium WebDriver implements support for various web browsers through driver classes, which provide driver specific functionality to the tests. The following list is of the major drivers supported by WebDriver:
Firefox
FirefoxDriver
is widely used and is a mature driver supporting most of the WebDriver core APIs including HTML5. Firefox driver is supported on major OS platforms including Windows, Linux, and Mac OS X.Language
Syntax
Java
WebDriver driver = new FirefoxDriver();
C#
IWebDriver driver = new FirefoxDriver();
Ruby
driver = Selenium::WebDriver.for :firefox
Python
driver = webdriver.Firefox()
For more information on
FirefoxDriver
, see http://code.google.com/p/selenium/wiki/FirefoxDriver.Google Chrome
ChromeDriver
is supported by installing a standaloneChromeDriver
server, which uses a JSON wire protocol to communicate between the Chrome browser and your test.chromeDriver
is supported on major OS platforms including Windows, Linux, and Mac OS X. Chrome has certain minor support limitations and does not support the HTML5 API.Language
Syntax
Java
WebDriver driver = new ChromeDriver();
C#
IWebDriver driver = new ChromeDriver();
Ruby
driver = Selenium::WebDriver.for :chrome
Python
driver = webdriver.Chrome()
For more information on
ChromeDriver
, see http://code.google.com/p/selenium/wiki/ChromeDriver.Internet Explorer
InternetExplorerDriver
is now a standalone server, which implements WebDriver's wire protocol. The driver supports running 32-bit and 64-bit versions of the browser on Windows.Language
Syntax
Java
WebDriver driver = new InternetExplorerDriver();
C#
IWebDriver driver = new InternetExplorerDriver();
Ruby
driver = Selenium::WebDriver.for :ie
Python
driver = webdriver.Ie()
For more information on
InternetExplorerDriver
, see http://code.google.com/p/selenium/wiki/InternetExplorerDriver.Safari
SafariDriver
is added to the list of supported browsers recently. It is implemented as a Safari browser extension instead of the client/server model used inChromeDriver
andInternetExplorerDriver
. It communicates with the WebDriver client using WebSockets supported on the Mac OS X and Windows platforms.Language
Syntax
Java
driver = new SafariDriver();
C#
IWebDriver driver = new SafariDriver();
Ruby
driver = Selenium::WebDriver.for :safari
Python
driver = webdriver.Remote("http://localhost:4444/wd/hub ",webdriver.DesiredCapabilities.SAFARI)
For information on
SafariDriver
, see http://code.google.com/p/selenium/wiki/SafariDriver.Opera
OperaDriver
is developed by Opera Software and its community. It is supported on all major OS platforms and is available as a core API and standalone server.Language
Syntax
Java
WebDriver driver = new OperaDriver();
C#
IWebDriver driver = new RemoteWebDriver(new Uri("http://localhost:4444/wd/hub
"), DesiredCapabilities.Opera());
Ruby
driver = Selenium::WebDriver.for :opera
Python
driver = webdriver.Opera()
For more information on
OperaDriver
, see http://code.google.com/p/selenium/wiki/OperaDriver.iPhone/iPad
IPhoneDriver
allows testing web applications on iOS using a special application that usesUIWebView
(a WebKit browser accessible for third-party applications) on iOS devices. This is done by installing and running a iWebDriver app on the iOS device or simulator.Language
Syntax
Java
WebDriver driver = new IPhoneDriver();
C#
IWebDriver driver = new RemoteWebDriver(new Uri("http://localhost:3001/wd/hub"), DesiredCapabilities.IPhone());
Ruby
driver = Selenium::WebDriver.for :iphone
Python
driver = webdriver.Remote("http://localhost:3001/wd/hub", webdriver.DesiredCapabilities.IPHONE)
For more information on
iPhoneDriver
, see http://code.google.com/p/selenium/wiki/IPhoneDriver.Android
AndroidDriver
allows testing of web applications on the Android browser. Similar toiPhoneDriver
, an Android application running on the device or emulator is used to run the tests.Language
Syntax
Java
WebDriver driver = new AndroidDriver();
C#
IWebDriver driver = AndroidDriver();
Ruby
driver = Selenium::WebDriver.for :android
Python
driver = webdriver.Remote("http://localhost:3001/wd/hub", webdriver.DesiredCapabilities.ANDROID)
For more information on
AndroidDriver
, see http://code.google.com/p/selenium/wiki/AndroidDriver.RemoteWebDriver
RemoteWebDriver
is a critical component of Selenium that allows testing on browsers located on remote machines. You can useRemoteWebDriver
for running your tests in a distributed architecture. TheRemoteWebDriver
consists of a client and server. The server is simply a Java Servlet running within the Jetty Servlet container. This servlet interacts with the various browsers. In the beginning of this section we sawselenium-server-standalone
, which is theRemoteWebDriver
server. The client is an instance ofRemoteWebDriver
, which communicates with the server via a JSON wire protocol similar to other drivers. We can specify what configuration is needed for testing by usingDesiredCapabilities
. For example we need to test an application on an iPad, and we can configure theRemoteWebDriver
in the following way to run the test:WebDriver driver = new RemoteWebDriver(new URL("http://localhost:3001/wd/hub"), DesiredCapabilities.ipad());
For more information on
RemoteWebDriver
, see http://code.google.com/p/selenium/wiki/RemoteWebDriver.
One of the key features of Selenium is its ability to locate different types of elements used on a page and the ability to interact with them. Selenium provides various ways or strategies to locate elements and perform actions such as clicking, typing text, selecting an option from the element, or performing verification on the state or property of the element, such as a text value. We can also check if the element is enabled or disabled, is checked or unchecked.
Selenium IDE: For locating elements using Selenium IDE, you need to specify locator details in the Target field.
Selenium WebDriver: For locating elements using WebDriver, the
driver
interface provides thefindElement()
andfindElements()
methods, which take locator expressions and search for the matching element(s).
The following table describes some of the important locator strategies that you can use with IDE or WebDriver API:
Strategy |
Description |
Selenium IDE |
Selenium WebDriver (Java) |
---|---|---|---|
|
Locates the first element that has the specified value for the |
|
|
|
Locates the first element with the specified value for the |
|
|
|
Locates the link (anchor) element that matches the specified pattern in link text. |
|
(You can specify a partial link text instead of complete link text. This might be useful when links partially contain dynamic values.) |
|
Locates elements using CSS selectors (refer to http://goo.gl/a9v7T and http://goo.gl/c1IRf for more details). CSS selectors are useful when the above selectors do not work to identify elements uniquely. You can also use XPath as an alternative, but CSS is considered faster with respect to strategy as compared to XPath. |
|
|
|
Locates elements using an XPath query. XPath is used to query XML documents and while HTML is subset of XML and browsers represent HTML documents as XHTML. We can use this strategy to locate elements. |
|
|
|
This strategy uses JavaScript expressions to find elements from the document object model (DOM). |
|
NA |
In addition to the aforementioned strategies, Selenium WebDriver supports the following strategies:
Strategy |
Description |
WebDriver API (Java) example |
---|---|---|
By class name |
Locates the first element that has the specified value for the |
|
By tag name |
Locates elements using their HTML tag |
For example, we can locate all the link (anchor) elements from a page with |
Note
Refer to Packt Publishing's Selenium Testing Tools Cookbook (http://www.packtpub.com/recipes-to-master-selenium-2-testing-tools-cookbook/book) for more information on locators.
Selenium provides an extensive support for the standard HTML elements used on a web page. You can interact with the HTML elements by using built-in commands and APIs for building simple to complex tests. The following table shows some key commands and API methods for interacting with a page and its elements:
Purpose |
Selenium IDE |
Selenium WebDriver (Java) |
---|---|---|
Click on an element |
|
|
Type a text |
|
|
Check a checkbox/radio button |
|
|
Uncheck Checkbox/Radio Button |
|
|
Select item(s) in a list or drop-down list |
|
|
Remove selection from a list or a drop-down list |
|
or
or
|
Get the inner text from the element |
|
|
Get attribute value |
- |
|
Selenium WebDriver provides the WebElement
class for interacting with HTML elements.
When Selenium scripts are played back, the application may not always respond with the same speed, especially for applications using AJAX. For example, it might take a few seconds for the following:
To load page contents
For a window or pop up message to open
For a progress bar to reach 100 percent
For a status message to appear
For a button to become enabled
You can handle these anticipated timing problems by synchronizing your script to ensure that Selenium waits until your application is ready before performing a certain step. There are several options that you can use to synchronize your script using Selenium IDE and Selenium WebDriver.
Selenium IDE provides various built-in waitFor
commands for handling synchronization problems in tests. The following is a is list of some waitFor
commands:
Command |
Condition |
---|---|
|
Delays execution until the page is fully loaded in the browser |
|
Delays execution until the specified element is present on the page |
|
Delays execution until the specified element is removed from the page |
|
Delays execution until the specified text is present on the page |
|
Delays execution until the contents of the frame are fully loaded in the browser |
|
Delays execution until an alert window is displayed on the page |
A number of these commands are run implicitly when other commands are being executed. For example, with respect to the clickAndWait
command, when you click on an element, the waitForPageToLoad
command is also executed.
Selenium WebDriver provides implicit and explicit wait conditions to handle synchronization problems. You can use these conditions in the following ways:
Option 1 – the implicit wait condition
When an implicit wait condition is implemented, if Selenium WebDriver cannot find an element in the DOM, it will wait for a defined amount of time for the element to appear in the DOM. In other terms, an implicit wait condition polls the DOM for a certain amount of time when trying to find an element(s) if it is not immediately available. The default setting is 0. Once set, the implicit wait condition is set for the life of the WebDriver object's instance. Here is a sample test using the implicit wait condition:
@Test public void testWithImplicitWait() { //Go to the Demo AJAX Application WebDriver driver = new FirefoxDriver(); driver.get("http://dl.dropbox.com/u/55228056/AjaxDemo.html"); //Set the Implicit Wait time Out to 10 Seconds driver.manage().timeouts().implicitlyWait(10, TimeUnit.SECONDS); try { //Get link for Page 4 and click on it WebElement page4button = driver.findElement(By.linkText("Page4")); page4button.click(); //Get an element with id page4 and verify it's text WebElement message = driver.findElement(By.id("page4")); assertTrue(message.getText().contains("Nunc nibh tortor")); } catch (NoSuchElementException e) { fail("Element not found!!"); e.printStackTrace(); } finally { driver.close(); } }
However, an implicit wait condition may slow down your tests when an application responds normally, as it will wait for each element appearing in the DOM and increase the overall execution time. It is recommended to avoid or minimize the use of an implicit wait condition.
Option 2 – the explicit wait condition
The explicit wait condition provides a better control compared with an implicit wait condition. Unlike an implicit wait condition, you can write custom code or conditions for a wait before proceeding further in the code. An explicit wait condition can only be implemented in cases where synchronization is needed and the rest of the script is working fine.
The Selenium WebDriver provides the
WebDriverWait
andExpectedCondition
classes for implementing an explicit wait condition. You can use these classes in the following way:@Test public void testExplcitWaitTitleContains() { //Go to the Google Home Page WebDriver driver = new FirefoxDriver(); driver.get("http://www.google.com"); //Enter a term to search and submit WebElement query = driver.findElement(By.name("q")); query.sendKeys("selenium"); query.click(); //Create Wait using WebDriverWait. //This will wait for 10 seconds for timeout before title is updated with search term //If title is updated in specified time limit test will move to the text step //instead of waiting for 10 seconds WebDriverWait wait = new WebDriverWait(driver, 10); wait.until(ExpectedConditions.titleContains("selenium")); //Verify Title assertTrue(driver.getTitle().toLowerCase(). startsWith("selenium")); driver.quit(); }
The
ExpectedCondition
class provides a set of predefined conditions to wait before proceeding further in the code. The following table shows some common conditions that we frequently come across when automating web browsers supported by theExpectedCondition
class:
Predefined condition |
Selenium WebDriver (Java) |
---|---|
An element is visible and enabled |
|
An element is selected |
|
Presence of an element |
|
Specific text present in an element |
|
Element value |
|
Title |
|
Note
For more conditions, visit http://selenium.googlecode.com/svn/trunk/docs/api/java/org/openqa/selenium/support/ui/ExpectedConditions.html.
The Page Object pattern provides tests for an interface, where a test can operate on the logical functionality offered by the page in a manner similar to the user accessing the page, but by hiding its internals. For example, if we build a Page Object for a login page that will provide a method to log in by accepting the username and password, and will take the user to the home page of the application. The test need not worry about what type of input controls are used for the login page, their locator details, navigation, and so on.
Tests should use objects of a page at a high level, where any change in layout or attributes used for the fields in the underlying page should not break the test.
Selenium WebDriver provides outstanding support for implementing the Page Object pattern via its PageFactory
class. The Page Object pattern brings the following advantages for your tests:
It helps in building a layer of abstraction separating automation code, which knows about locating application elements and the one which interacts with these elements for actual testing
It provides a central repository of pages from the application for tests
It provides high maintainability and reduction in code duplication
The login page class provides an interface to the login page of the application to the tests as shown in the following diagram:
Here is Page Object for the login page of the test application.
The login page class contains locator details for the key elements needed for user login functionality. Elements from the login page are defined as private members of the LoginPage
class. The test code will not have access to these elements. The LoginPage
class provides a public login()
method to the tests. The test needs to pass the e-mail address and password for a registered user to this method. The constructor of the LoginPage
class uses the PageFactory.initElements()
method to initialize WebElements defined in the class in the following way:
package demo.magentocommerce.pages; import org.openqa.selenium.WebDriver; import org.openqa.selenium.WebElement; import org.openqa.selenium.support.FindBy; import org.openqa.selenium.support.PageFactory; public class Login { @FindBy(id="email") private WebElement emailField; @FindBy(id="pass") private WebElement passwordField; @FindBy(id="send2") private WebElement loginButton; public Login(WebDriver driver) { PageFactory.initElements(driver, this); } public void login(String email, String password) { emailField.sendKeys(email); passwordField.sendKeys(email); loginButton.click(); } }
Here is a sample test using the LoginPage
class. The instance of the LoginPage
class is created by passing the current driver
instance. The login()
method is called in the following way:
... //Navigate to the Login Page WebDriver driver = new FirefoxDriver(); driver.get("http://demo.magentocommerce.com/customer/account/login/"); //Create an instance of Login page and call the Login function LoginPage loginPage = new LoginPage(driver); loginPage.login("[email protected]","xxxx"); //Verify the user is logged in ...