Legacy

Documentation related to the legacy components of Selenium. Meant to be kept purely for historical reasons and not as a incentive to use deprecated components.

Most of the documentation found in this section is still in English. Please note we are not accepting pull requests to translate this content as translating documentation of legacy components does not add value to the community nor the project.

1 - Selenium 1 (Selenium RC)

Most of the documentation found in this section is still in English. Please note we are not accepting pull requests to translate this content as translating documentation of legacy components does not add value to the community nor the project.

Introduction

Selenium RC was the main Selenium project for a long time, before the WebDriver/Selenium merge brought up Selenium 2, a more powerful tool. It is worth to highlight that Selenium 1 is not supported anymore.

How Selenium RC Works

First, we will describe how the components of Selenium RC operate and the role each plays in running your test scripts.

RC Components

Selenium RC components are:

  • The Selenium Server which launches and kills browsers, interprets and runs the Selenese commands passed from the test program, and acts as an HTTP proxy, intercepting and verifying HTTP messages passed between the browser and the AUT.
  • Client libraries which provide the interface between each programming language and the Selenium RC Server.

Here is a simplified architecture diagram:

Architecture Diagram Simple

The diagram shows the client libraries communicate with the Server passing each Selenium command for execution. Then the server passes the Selenium command to the browser using Selenium-Core JavaScript commands. The browser, using its JavaScript interpreter, executes the Selenium command. This runs the Selenese action or verification you specified in your test script.

Selenium Server

Selenium Server receives Selenium commands from your test program, interprets them, and reports back to your program the results of running those tests.

The RC server bundles Selenium Core and automatically injects it into the browser. This occurs when your test program opens the browser (using a client library API function). Selenium-Core is a JavaScript program, actually a set of JavaScript functions which interprets and executes Selenese commands using the browser’s built-in JavaScript interpreter.

The Server receives the Selenese commands from your test program using simple HTTP GET/POST requests. This means you can use any programming language that can send HTTP requests to automate Selenium tests on the browser.

Client Libraries

The client libraries provide the programming support that allows you to run Selenium commands from a program of your own design. There is a different client library for each supported language. A Selenium client library provides a programming interface (API), i.e., a set of functions, which run Selenium commands from your own program. Within each interface, there is a programming function that supports each Selenese command.

The client library takes a Selenese command and passes it to the Selenium Server for processing a specific action or test against the application under test (AUT). The client library also receives the result of that command and passes it back to your program. Your program can receive the result and store it into a program variable and report it as a success or failure, or possibly take corrective action if it was an unexpected error.

So to create a test program, you simply write a program that runs a set of Selenium commands using a client library API. And, optionally, if you already have a Selenese test script created in the Selenium-IDE, you can generate the Selenium RC code. The Selenium-IDE can translate (using its Export menu item) its Selenium commands into a client-driver’s API function calls. See the Selenium-IDE chapter for specifics on exporting RC code from Selenium-IDE.

Installation

Installation is rather a misnomer for Selenium. Selenium has a set of libraries available in the programming language of your choice. You could download them from the downloads page.

Once you’ve chosen a language to work with, you simply need to:

  • Install the Selenium RC Server.
  • Set up a programming project using a language specific client driver.

Installing Selenium Server

The Selenium RC server is simply a Java jar file (selenium-server-standalone-.jar), which doesn’t require any special installation. Just downloading the zip file and extracting the server in the desired directory is sufficient.

Running Selenium Server

Before starting any tests you must start the server. Go to the directory where Selenium RC’s server is located and run the following from a command-line console.

    java -jar selenium-server-standalone-<version-number>.jar

This can be simplified by creating a batch or shell executable file (.bat on Windows and .sh on Linux) containing the command above. Then make a shortcut to that executable on your desktop and simply double-click the icon to start the server.

For the server to run you’ll need Java installed and the PATH environment variable correctly configured to run it from the console. You can check that you have Java correctly installed by running the following on a console.

       java -version

If you get a version number (which needs to be 1.5 or later), you’re ready to start using Selenium RC.

Using the Java Client Driver

  • Download Selenium java client driver zip from the SeleniumHQ downloads page.
  • Extract selenium-java-.jar file
  • Open your desired Java IDE (Eclipse, NetBeans, IntelliJ, Netweaver, etc.)
  • Create a java project.
  • Add the selenium-java-.jar files to your project as references.
  • Add to your project classpath the file selenium-java-.jar.
  • From Selenium-IDE, export a script to a Java file and include it in your Java project, or write your Selenium test in Java using the selenium-java-client API. The API is presented later in this chapter. You can either use JUnit, or TestNg to run your test, or you can write your own simple main() program. These concepts are explained later in this section.
  • Run Selenium server from the console.
  • Execute your test from the Java IDE or from the command-line.

For details on Java test project configuration, see the Appendix sections Configuring Selenium RC With Eclipse and Configuring Selenium RC With Intellij.

Using the Python Client Driver

  • Install Selenium via PIP, instructions linked at SeleniumHQ downloads page
  • Either write your Selenium test in Python or export a script from Selenium-IDE to a python file.
  • Run Selenium server from the console
  • Execute your test from a console or your Python IDE

For details on Python client driver configuration, see the appendix Python Client Driver Configuration.

Using the .NET Client Driver

  • Download Selenium RC from the SeleniumHQ downloads page
  • Extract the folder
  • Download and install NUnit ( Note: You can use NUnit as your test engine. If you’re not familiar yet with NUnit, you can also write a simple main() function to run your tests; however NUnit is very useful as a test engine.)
  • Open your desired .Net IDE (Visual Studio, SharpDevelop, MonoDevelop)
  • Create a class library (.dll)
  • Add references to the following DLLs: nmock.dll, nunit.core.dll, nunit. framework.dll, ThoughtWorks.Selenium.Core.dll, ThoughtWorks.Selenium.IntegrationTests.dll and ThoughtWorks.Selenium.UnitTests.dll
  • Write your Selenium test in a .Net language (C#, VB.Net), or export a script from Selenium-IDE to a C# file and copy this code into the class file you just created.
  • Write your own simple main() program or you can include NUnit in your project for running your test. These concepts are explained later in this chapter.
  • Run Selenium server from console
  • Run your test either from the IDE, from the NUnit GUI or from the command line

For specific details on .NET client driver configuration with Visual Studio, see the appendix .NET client driver configuration.

Using the Ruby Client Driver

  • If you do not already have RubyGems, install it from RubyForge.
  • Run gem install selenium-client
  • At the top of your test script, add require "selenium/client"
  • Write your test script using any Ruby test harness (eg Test::Unit, Mini::Test or RSpec).
  • Run Selenium RC server from the console.
  • Execute your test in the same way you would run any other Ruby script.

For details on Ruby client driver configuration, see the Selenium-Client documentation_

From Selenese to a Program

The primary task for using Selenium RC is to convert your Selenese into a programming language. In this section, we provide several different language-specific examples.

Sample Test Script

Let’s start with an example Selenese test script. Imagine recording the following test with Selenium-IDE.

open/
typeqselenium rc
clickAndWaitbtnG
assertTextPresentResults * for selenium rc

Note: This example would work with the Google search page http://www.google.com

Selenese as Programming Code

Here is the test script exported (via Selenium-IDE) to each of the supported programming languages. If you have at least basic knowledge of an object- oriented programming language, you will understand how Selenium runs Selenese commands by reading one of these examples. To see an example in a specific language, select one of these buttons.

CSharp


        using System;
        using System.Text;
        using System.Text.RegularExpressions;
        using System.Threading;
        using NUnit.Framework;
        using Selenium;

        namespace SeleniumTests
        {
            [TestFixture]
            public class NewTest
            {
                private ISelenium selenium;
                private StringBuilder verificationErrors;
                
                [SetUp]
                public void SetupTest()
                {
                    selenium = new DefaultSelenium("localhost", 4444, "*firefox", "http://www.google.com/");
                    selenium.Start();
                    verificationErrors = new StringBuilder();
                }
                
                [TearDown]
                public void TeardownTest()
                {
                    try
                    {
                        selenium.Stop();
                    }
                    catch (Exception)
                    {
                        // Ignore errors if unable to close the browser
                    }
                    Assert.AreEqual("", verificationErrors.ToString());
                }
                
                [Test]
                public void TheNewTest()
                {
                    selenium.Open("/");
                    selenium.Type("q", "selenium rc");
                    selenium.Click("btnG");
                    selenium.WaitForPageToLoad("30000");
                    Assert.AreEqual("selenium rc - Google Search", selenium.GetTitle());
                }
            }
        }

Java

      
	  /** Add JUnit framework to your classpath if not already there 
	   *  for this example to work
	  */
      package com.example.tests;

      import com.thoughtworks.selenium.*;
      import java.util.regex.Pattern;

      public class NewTest extends SeleneseTestCase {
          public void setUp() throws Exception {
              setUp("http://www.google.com/", "*firefox");
          }
            public void testNew() throws Exception {
                selenium.open("/");
                selenium.type("q", "selenium rc");
                selenium.click("btnG");
                selenium.waitForPageToLoad("30000");
                assertTrue(selenium.isTextPresent("Results * for selenium rc"));
          }
      }

Php

      <?php

      require_once 'PHPUnit/Extensions/SeleniumTestCase.php';

      class Example extends PHPUnit_Extensions_SeleniumTestCase
      {
        function setUp()
        {
          $this->setBrowser("*firefox");
          $this->setBrowserUrl("http://www.google.com/");
        }

        function testMyTestCase()
        {
          $this->open("/");
          $this->type("q", "selenium rc");
          $this->click("btnG");
          $this->waitForPageToLoad("30000");
          $this->assertTrue($this->isTextPresent("Results * for selenium rc"));
        }
      }
      ?>

Python


     from selenium import selenium
      import unittest, time, re

      class NewTest(unittest.TestCase):
          def setUp(self):
              self.verificationErrors = []
              self.selenium = selenium("localhost", 4444, "*firefox",
                      "http://www.google.com/")
              self.selenium.start()
         
          def test_new(self):
              sel = self.selenium
              sel.open("/")
              sel.type("q", "selenium rc")
              sel.click("btnG")
              sel.wait_for_page_to_load("30000")
              self.failUnless(sel.is_text_present("Results * for selenium rc"))
         
          def tearDown(self):
              self.selenium.stop()
              self.assertEqual([], self.verificationErrors)

Ruby


      require "selenium/client"
      require "test/unit"

      class NewTest < Test::Unit::TestCase
        def setup
          @verification_errors = []
          if $selenium
            @selenium = $selenium
          else
            @selenium =  Selenium::Client::Driver.new("localhost", 4444, "*firefox", "http://www.google.com/", 60);
            @selenium.start
          end
          @selenium.set_context("test_new")
        end

        def teardown
          @selenium.stop unless $selenium
          assert_equal [], @verification_errors
        end

        def test_new
          @selenium.open "/"
          @selenium.type "q", "selenium rc"
          @selenium.click "btnG"
          @selenium.wait_for_page_to_load "30000"
          assert @selenium.is_text_present("Results * for selenium rc")
        end
      end

In the next section we’ll explain how to build a test program using the generated code.

Programming Your Test

Now we’ll illustrate how to program your own tests using examples in each of the supported programming languages. There are essentially two tasks:

  • Generate your script into a programming language from Selenium-IDE, optionally modifying the result.
  • Write a very simple main program that executes the generated code.

Optionally, you can adopt a test engine platform like JUnit or TestNG for Java, or NUnit for .NET if you are using one of those languages.

Here, we show language-specific examples. The language-specific APIs tend to differ from one to another, so you’ll find a separate explanation for each.

  • Java
  • C#
  • Python
  • Ruby
  • Perl, PHP

Java

For Java, people use either JUnit or TestNG as the test engine.
Some development environments like Eclipse have direct support for these via plug-ins. This makes it even easier. Teaching JUnit or TestNG is beyond the scope of this document however materials may be found online and there are publications available. If you are already a “java-shop” chances are your developers will already have some experience with one of these test frameworks.

You will probably want to rename the test class from “NewTest” to something of your own choosing. Also, you will need to change the browser-open parameters in the statement:

    selenium = new DefaultSelenium("localhost", 4444, "*iehta", "http://www.google.com/");

The Selenium-IDE generated code will look like this. This example has comments added manually for additional clarity.

   package com.example.tests;
   // We specify the package of our tests

   import com.thoughtworks.selenium.*;
   // This is the driver's import. You'll use this for instantiating a
   // browser and making it do what you need.

   import java.util.regex.Pattern;
   // Selenium-IDE add the Pattern module because it's sometimes used for 
   // regex validations. You can remove the module if it's not used in your 
   // script.

   public class NewTest extends SeleneseTestCase {
   // We create our Selenium test case

         public void setUp() throws Exception {
           setUp("http://www.google.com/", "*firefox");
                // We instantiate and start the browser
         }

         public void testNew() throws Exception {
              selenium.open("/");
              selenium.type("q", "selenium rc");
              selenium.click("btnG");
              selenium.waitForPageToLoad("30000");
              assertTrue(selenium.isTextPresent("Results * for selenium rc"));
              // These are the real test steps
        }
   }

C#

The .NET Client Driver works with Microsoft.NET. It can be used with any .NET testing framework like NUnit or the Visual Studio 2005 Team System.

Selenium-IDE assumes you will use NUnit as your testing framework. You can see this in the generated code below. It includes the using statement for NUnit along with corresponding NUnit attributes identifying the role for each member function of the test class.

You will probably have to rename the test class from “NewTest” to something of your own choosing. Also, you will need to change the browser-open parameters in the statement:

    selenium = new DefaultSelenium("localhost", 4444, "*iehta", "http://www.google.com/");

The generated code will look similar to this.


    using System;
    using System.Text;
    using System.Text.RegularExpressions;
    using System.Threading;
    using NUnit.Framework;
    using Selenium;
    
    namespace SeleniumTests

    {
        [TestFixture]

        public class NewTest

        {
        private ISelenium selenium;

        private StringBuilder verificationErrors;

        [SetUp]

        public void SetupTest()

        {
            selenium = new DefaultSelenium("localhost", 4444, "*iehta",
            "http://www.google.com/");

            selenium.Start();

            verificationErrors = new StringBuilder();
        }

        [TearDown]

        public void TeardownTest()
        {
            try
            {
            selenium.Stop();
            }

            catch (Exception)
            {
            // Ignore errors if unable to close the browser
            }

            Assert.AreEqual("", verificationErrors.ToString());
        }
        [Test]

        public void TheNewTest()
        {
            // Open Google search engine.        
            selenium.Open("http://www.google.com/"); 
            
            // Assert Title of page.
            Assert.AreEqual("Google", selenium.GetTitle());
            
            // Provide search term as "Selenium OpenQA"
            selenium.Type("q", "Selenium OpenQA");
            
            // Read the keyed search term and assert it.
            Assert.AreEqual("Selenium OpenQA", selenium.GetValue("q"));
            
            // Click on Search button.
            selenium.Click("btnG");
            
            // Wait for page to load.
            selenium.WaitForPageToLoad("5000");
            
            // Assert that "www.openqa.org" is available in search results.
            Assert.IsTrue(selenium.IsTextPresent("www.openqa.org"));
            
            // Assert that page title is - "Selenium OpenQA - Google Search"
            Assert.AreEqual("Selenium OpenQA - Google Search", 
                         selenium.GetTitle());
        }
        }
    }

You can allow NUnit to manage the execution of your tests. Or alternatively, you can write a simple main() program that instantiates the test object and runs each of the three methods, SetupTest(), TheNewTest(), and TeardownTest() in turn.

Python

Pyunit is the test framework to use for Python.

The basic test structure is:


   from selenium import selenium
   # This is the driver's import.  You'll use this class for instantiating a
   # browser and making it do what you need.

   import unittest, time, re
   # This are the basic imports added by Selenium-IDE by default.
   # You can remove the modules if they are not used in your script.

   class NewTest(unittest.TestCase):
   # We create our unittest test case

       def setUp(self):
           self.verificationErrors = []
           # This is an empty array where we will store any verification errors
           # we find in our tests

           self.selenium = selenium("localhost", 4444, "*firefox",
                   "http://www.google.com/")
           self.selenium.start()
           # We instantiate and start the browser

       def test_new(self):
           # This is the test code.  Here you should put the actions you need
           # the browser to do during your test.
            
           sel = self.selenium
           # We assign the browser to the variable "sel" (just to save us from 
           # typing "self.selenium" each time we want to call the browser).
            
           sel.open("/")
           sel.type("q", "selenium rc")
           sel.click("btnG")
           sel.wait_for_page_to_load("30000")
           self.failUnless(sel.is_text_present("Results * for selenium rc"))
           # These are the real test steps

       def tearDown(self):
           self.selenium.stop()
           # we close the browser (I'd recommend you to comment this line while
           # you are creating and debugging your tests)

           self.assertEqual([], self.verificationErrors)
           # And make the test fail if we found that any verification errors
           # were found

Ruby

Old (pre 2.0) versions of Selenium-IDE generate Ruby code that requires the old Selenium gem. Therefore, it is advisable to update any Ruby scripts generated by the IDE as follows:

  1. On line 1, change require "selenium" to require "selenium/client"

  2. On line 11, change Selenium::SeleniumDriver.new to Selenium::Client::Driver.new

You probably also want to change the class name to something more informative than “Untitled,” and change the test method’s name to something other than “test_untitled.”

Here is a simple example created by modifying the Ruby code generated by Selenium IDE, as described above.


   # load the Selenium-Client gem
   require "selenium/client"

   # Load Test::Unit, Ruby's default test framework.
   # If you prefer RSpec, see the examples in the Selenium-Client
   # documentation.
   require "test/unit"

   class Untitled < Test::Unit::TestCase

     # The setup method is called before each test.
     def setup

       # This array is used to capture errors and display them at the
       # end of the test run.
       @verification_errors = []

       # Create a new instance of the Selenium-Client driver.
       @selenium = Selenium::Client::Driver.new \
         :host => "localhost",
         :port => 4444,
         :browser => "*chrome",
         :url => "http://www.google.com/",
         :timeout_in_second => 60

       # Start the browser session
       @selenium.start

       # Print a message in the browser-side log and status bar
       # (optional).
       @selenium.set_context("test_untitled")

     end

     # The teardown method is called after each test.
     def teardown

       # Stop the browser session.
       @selenium.stop

       # Print the array of error messages, if any.
       assert_equal [], @verification_errors
     end

     # This is the main body of your test.
     def test_untitled
     
       # Open the root of the site we specified when we created the
       # new driver instance, above.
       @selenium.open "/"

       # Type 'selenium rc' into the field named 'q'
       @selenium.type "q", "selenium rc"

       # Click the button named "btnG"
       @selenium.click "btnG"

       # Wait for the search results page to load.
       # Note that we don't need to set a timeout here, because that
       # was specified when we created the new driver instance, above.
       @selenium.wait_for_page_to_load

       begin

          # Test whether the search results contain the expected text.
	  # Notice that the star (*) is a wildcard that matches any
	  # number of characters.
	  assert @selenium.is_text_present("Results * for selenium rc")
	  
       rescue Test::Unit::AssertionFailedError
       
          # If the assertion fails, push it onto the array of errors.
	  @verification_errors << $!

       end
     end
   end

Perl, PHP

The members of the documentation team have not used Selenium RC with Perl or PHP. If you are using Selenium RC with either of these two languages please contact the Documentation Team (see the chapter on contributing). We would love to include some examples from you and your experiences, to support Perl and PHP users.

Learning the API

The Selenium RC API uses naming conventions that, assuming you understand Selenese, much of the interface
will be self-explanatory. Here, however, we explain the most critical and possibly less obvious aspects.

Starting the Browser

CSharp

      selenium = new DefaultSelenium("localhost", 4444, "*firefox", "http://www.google.com/");
      selenium.Start();

Java


      setUp("http://www.google.com/", "*firefox");

Perl

      my $sel = Test::WWW::Selenium->new( host => "localhost", 
                                          port => 4444, 
                                          browser => "*firefox", 
                                          browser_url => "http://www.google.com/" );

Php

      $this->setBrowser("*firefox");
      $this->setBrowserUrl("http://www.google.com/");

Python

      self.selenium = selenium("localhost", 4444, "*firefox",
                               "http://www.google.com/")
      self.selenium.start()

Ruby

      @selenium = Selenium::ClientDriver.new("localhost", 4444, "*firefox", "http://www.google.com/", 10000);
      @selenium.start

Each of these examples opens the browser and represents that browser by assigning a “browser instance” to a program variable. This program variable is then used to call methods from the browser. These methods execute the Selenium commands, i.e. like open or type or the verify commands.

The parameters required when creating the browser instance are:

  • host Specifies the IP address of the computer where the server is located. Usually, this is the same machine as where the client is running, so in this case localhost is passed. In some clients this is an optional parameter.

  • port Specifies the TCP/IP socket where the server is listening waiting for the client to establish a connection. This also is optional in some client drivers.

  • browser The browser in which you want to run the tests. This is a required parameter.

  • url The base url of the application under test. This is required by all the client libs and is integral information for starting up the browser-proxy-AUT communication.

Note that some of the client libraries require the browser to be started explicitly by calling its start() method.

Running Commands

Once you have the browser initialized and assigned to a variable (generally named “selenium”) you can make it run Selenese commands by calling the respective methods from the browser variable. For example, to call the type method of the selenium object:

    selenium.type("field-id","string to type")

In the background the browser will actually perform a type operation, essentially identical to a user typing input into the browser, by
using the locator and the string you specified during the method call.

Reporting Results

Selenium RC does not have its own mechanism for reporting results. Rather, it allows you to build your reporting customized to your needs using features of your chosen programming language. That’s great, but what if you simply want something quick that’s already done for you? Often an existing library or test framework can meet your needs faster than developing your own test reporting code.

Test Framework Reporting Tools

Test frameworks are available for many programming languages. These, along with their primary function of providing a flexible test engine for executing your tests, include library code for reporting results. For example, Java has two commonly used test frameworks, JUnit and TestNG. .NET also has its own, NUnit.

We won’t teach the frameworks themselves here; that’s beyond the scope of this user guide. We will simply introduce the framework features that relate to Selenium along with some techniques you can apply. There are good books available on these test frameworks however along with information on the internet.

Test Report Libraries

Also available are third-party libraries specifically created for reporting test results in your chosen programming language. These often support a variety of formats such as HTML or PDF.

What’s The Best Approach?

Most people new to the testing frameworks will begin with the framework’s built-in reporting features. From there most will examine any available libraries as that’s less time consuming than developing your own. As you begin to use Selenium no doubt you will start putting in your own “print statements” for reporting progress. That may gradually lead to you developing your own reporting, possibly in parallel to using a library or test framework. Regardless, after the initial, but short, learning curve you will naturally develop what works best for your own situation.

Test Reporting Examples

To illustrate, we’ll direct you to some specific tools in some of the other languages supported by Selenium. The ones listed here are commonly used and have been used extensively (and therefore recommended) by the authors of this guide.

Test Reports in Java

  • If Selenium Test cases are developed using JUnit then JUnit Report can be used to generate test reports.

  • If Selenium Test cases are developed using TestNG then no external task is required to generate test reports. The TestNG framework generates an HTML report which list details of tests.

  • ReportNG is a HTML reporting plug-in for the TestNG framework. It is intended as a replacement for the default TestNG HTML report. ReportNG provides a simple, colour-coded view of the test results.

Logging the Selenese Commands
  • Logging Selenium can be used to generate a report of all the Selenese commands in your test along with the success or failure of each. Logging Selenium extends the Java client driver to add this Selenese logging ability.

Test Reports for Python

  • When using Python Client Driver then HTMLTestRunner can be used to generate a Test Report.

Test Reports for Ruby

  • If RSpec framework is used for writing Selenium Test Cases in Ruby then its HTML report can be used to generate a test report.

Adding Some Spice to Your Tests

Now we’ll get to the whole reason for using Selenium RC, adding programming logic to your tests. It’s the same as for any program. Program flow is controlled using condition statements and iteration. In addition you can report progress information using I/O. In this section we’ll show some examples of how programming language constructs can be combined with Selenium to solve common testing problems.

You will find as you transition from the simple tests of the existence of page elements to tests of dynamic functionality involving multiple web-pages and varying data that you will require programming logic for verifying expected results. Basically, the Selenium-IDE does not support iteration and standard condition statements. You can do some conditions by embedding JavaScript in Selenese parameters, however iteration is impossible, and most conditions will be much easier in a
programming language. In addition, you may need exception handling for error recovery. For these reasons and others, we have written this section to illustrate the use of common programming techniques to give you greater ‘verification power’ in your automated testing.

The examples in this section are written in C# and Java, although the code is simple and can be easily adapted to the other supported languages. If you have some basic knowledge of an object-oriented programming language you shouldn’t have difficulty understanding this section.

Iteration

Iteration is one of the most common things people need to do in their tests. For example, you may want to to execute a search multiple times. Or, perhaps for verifying your test results you need to process a “result set” returned from a database.

Using the same Google search example we used earlier, let’s check the Selenium search results. This test could use the Selenese:

open/
typeqselenium rc
clickAndWaitbtnG
assertTextPresentResults * for selenium rc
typeqselenium ide
clickAndWaitbtnG
assertTextPresentResults * for selenium ide
typeqselenium grid
clickAndWaitbtnG
assertTextPresentResults * for selenium grid

The code has been repeated to run the same steps 3 times. But multiple copies of the same code is not good program practice because it’s more work to maintain. By using a programming language, we can iterate over the search results for a more flexible and maintainable solution.

In C#

   // Collection of String values.
   String[] arr = {"ide", "rc", "grid"};    
        
   // Execute loop for each String in array 'arr'.
   foreach (String s in arr) {
       sel.open("/");
       sel.type("q", "selenium " +s);
       sel.click("btnG");
       sel.waitForPageToLoad("30000");
       assertTrue("Expected text: " +s+ " is missing on page."
       , sel.isTextPresent("Results * for selenium " + s));
    }

Condition Statements

To illustrate using conditions in tests we’ll start with an example. A common problem encountered while running Selenium tests occurs when an expected element is not available on page. For example, when running the following line:

   selenium.type("q", "selenium " +s);

If element ‘q’ is not on the page then an exception is thrown:

   com.thoughtworks.selenium.SeleniumException: ERROR: Element q not found

This can cause your test to abort. For some tests that’s what you want. But often that is not desirable as your test script has many other subsequent tests to perform.

A better approach is to first validate whether the element is really present and then take alternatives when it it is not. Let’s look at this using Java.

   // If element is available on page then perform type operation.
   if(selenium.isElementPresent("q")) {
       selenium.type("q", "Selenium rc");
   } else {
       System.out.printf("Element: " +q+ " is not available on page.")
   }

The advantage of this approach is to continue with test execution even if some UI elements are not available on page.

Executing JavaScript from Your Test

JavaScript comes very handy in exercising an application which is not directly supported by Selenium. The getEval method of Selenium API can be used to execute JavaScript from Selenium RC.

Consider an application having check boxes with no static identifiers. In this case one could evaluate JavaScript from Selenium RC to get ids of all check boxes and then exercise them.

   public static String[] getAllCheckboxIds () { 
		String script = "var inputId  = new Array();";// Create array in java script.
		script += "var cnt = 0;"; // Counter for check box ids.  
		script += "var inputFields  = new Array();"; // Create array in java script.
		script += "inputFields = window.document.getElementsByTagName('input');"; // Collect input elements.
		script += "for(var i=0; i<inputFields.length; i++) {"; // Loop through the collected elements.
		script += "if(inputFields[i].id !=null " +
		"&& inputFields[i].id !='undefined' " +
		"&& inputFields[i].getAttribute('type') == 'checkbox') {"; // If input field is of type check box and input id is not null.
		script += "inputId[cnt]=inputFields[i].id ;" + // Save check box id to inputId array.
		"cnt++;" + // increment the counter.
		"}" + // end of if.
		"}"; // end of for.
		script += "inputId.toString();" ;// Convert array in to string.			
		String[] checkboxIds = selenium.getEval(script).split(","); // Split the string.
		return checkboxIds;
    }

To count number of images on a page:

   selenium.getEval("window.document.images.length;");

Remember to use window object in case of DOM expressions as by default selenium window is referred to, not the test window.

Server Options

When the server is launched, command line options can be used to change the default server behaviour.

Recall, the server is started by running the following.

   $ java -jar selenium-server-standalone-<version-number>.jar

To see the list of options, run the server with the -h option.

   $ java -jar selenium-server-standalone-<version-number>.jar -h

You’ll see a list of all the options you can use with the server and a brief description of each. The provided descriptions will not always be enough, so we’ve provided explanations for some of the more important options.

Proxy Configuration

If your AUT is behind an HTTP proxy which requires authentication then you should configure http.proxyHost, http.proxyPort, http.proxyUser and http.proxyPassword using the following command.

   $ java -jar selenium-server-standalone-<version-number>.jar -Dhttp.proxyHost=proxy.com -Dhttp.proxyPort=8080 -Dhttp.proxyUser=username -Dhttp.proxyPassword=password

Multi-Window Mode

If you are using Selenium 1.0 you can probably skip this section, since multiwindow mode is the default behavior. However, prior to version 1.0, Selenium by default ran the application under test in a sub frame as shown here.

Single window mode

Some applications didn’t run correctly in a sub frame, and needed to be loaded into the top frame of the window. The multi-window mode option allowed the AUT to run in a separate window rather than in the default frame where it could then have the top frame it required.

Multiwindow Mode

For older versions of Selenium you must specify multiwindow mode explicitly with the following option:

   -multiwindow 

As of Selenium RC 1.0, if you want to run your test within a single frame (i.e. using the standard for earlier Selenium versions) you can state this to the Selenium Server using the option

   -singlewindow 

Specifying the Firefox Profile

Firefox will not run two instances simultaneously unless you specify a separate profile for each instance. Selenium RC 1.0 and later runs in a separate profile automatically, so if you are using Selenium 1.0, you can probably skip this section. However, if you’re using an older version of Selenium or if you need to use a specific profile for your tests (such as adding an https certificate or having some addons installed), you will need to explicitly specify the profile.

First, to create a separate Firefox profile, follow this procedure. Open the Windows Start menu, select “Run”, then type and enter one of the following:

   firefox.exe -profilemanager 
   firefox.exe -P 

Create the new profile using the dialog. Then when you run Selenium Server, tell it to use this new Firefox profile with the server command-line option -firefoxProfileTemplate and specify the path to the profile using its filename and directory path.

   -firefoxProfileTemplate "path to the profile" 

Warning: Be sure to put your profile in a new folder separate from the default!!! The Firefox profile manager tool will delete all files in a folder if you delete a profile, regardless of whether they are profile files or not.

More information about Firefox profiles can be found in Mozilla’s Knowledge Base

Run Selenese Directly Within the Server Using -htmlSuite

You can run Selenese html files directly within the Selenium Server by passing the html file to the server’s command line. For instance:

   java -jar selenium-server-standalone-<version-number>.jar -htmlSuite "*firefox" 
   "http://www.google.com" "c:\absolute\path\to\my\HTMLSuite.html" 
   "c:\absolute\path\to\my\results.html"

This will automatically launch your HTML suite, run all the tests and save a nice HTML report with the results.

Note: When using this option, the server will start the tests and wait for a specified number of seconds for the test to complete; if the test doesn’t complete within that amount of time, the command will exit with a non-zero exit code and no results file will be generated.

This command line is very long so be careful when you type it. Note this requires you to pass in an HTML Selenese suite, not a single test. Also be aware the -htmlSuite option is incompatible with -interactive You cannot run both at the same time.

Selenium Server Logging

Server-Side Logs

When launching Selenium server the -log option can be used to record valuable debugging information reported by the Selenium Server to a text file.

   java -jar selenium-server-standalone-<version-number>.jar -log selenium.log

This log file is more verbose than the standard console logs (it includes DEBUG level logging messages). The log file also includes the logger name, and the ID number of the thread that logged the message. For example:

   20:44:25 DEBUG [12] org.openqa.selenium.server.SeleniumDriverResourceHandler - 
   Browser 465828/:top frame1 posted START NEW

The message format is

   TIMESTAMP(HH:mm:ss) LEVEL [THREAD] LOGGER - MESSAGE

This message may be multiline.

Browser-Side Logs

JavaScript on the browser side (Selenium Core) also logs important messages; in many cases, these can be more useful to the end-user than the regular Selenium Server logs. To access browser-side logs, pass the -browserSideLog argument to the Selenium Server.

   java -jar selenium-server-standalone-<version-number>.jar -browserSideLog

-browserSideLog must be combined with the -log argument, to log browserSideLogs (as well as all other DEBUG level logging messages) to a file.

Specifying the Path to a Specific Browser

You can specify to Selenium RC a path to a specific browser. This is useful if you have different versions of the same browser and you wish to use a specific one. Also, this is used to allow your tests to run against a browser not directly supported by Selenium RC. When specifying the run mode, use the *custom specifier followed by the full path to the browser’s executable:

   *custom <path to browser> 

Selenium RC Architecture

Note: This topic tries to explain the technical implementation behind Selenium RC. It’s not fundamental for a Selenium user to know this, but could be useful for understanding some of the problems you might find in the future.

To understand in detail how Selenium RC Server works and why it uses proxy injection and heightened privilege modes you must first understand the same origin policy_.

The Same Origin Policy

The main restriction that Selenium faces is the Same Origin Policy. This security restriction is applied by every browser in the market and its objective is to ensure that a site’s content will never be accessible by a script from another site. The Same Origin Policy dictates that any code loaded within the browser can only operate within that website’s domain. It cannot perform functions on another website. So for example, if the browser loads JavaScript code when it loads www.mysite.com, it cannot run that loaded code against www.mysite2.com–even if that’s another of your sites. If this were possible, a script placed on any website you open would be able to read information on your bank account if you had the account page opened on other tab. This is called XSS (Cross-site Scripting).

To work within this policy, Selenium-Core (and its JavaScript commands that make all the magic happen) must be placed in the same origin as the Application Under Test (same URL).

Historically, Selenium-Core was limited by this problem since it was implemented in JavaScript. Selenium RC is not, however, restricted by the Same Origin Policy. Its use of the Selenium Server as a proxy avoids this problem. It, essentially, tells the browser that the browser is working on a single “spoofed” website that the Server provides.

Note: You can find additional information about this topic on Wikipedia pages about Same Origin Policy and XSS.

Proxy Injection

The first method Selenium used to avoid the The Same Origin Policy was Proxy Injection. In Proxy Injection Mode, the Selenium Server acts as a client-configured HTTP proxy1, that sits between the browser and the Application Under Test2. It then masks the AUT under a fictional URL (embedding Selenium-Core and the set of tests and delivering them as if they were coming from the same origin).

Here is an architectural diagram.

Architectural Diagram 1

As a test suite starts in your favorite language, the following happens:

  1. The client/driver establishes a connection with the selenium-RC server.
  2. Selenium RC server launches a browser (or reuses an old one) with a URL that injects Selenium-Core’s JavaScript into the browser-loaded web page.
  3. The client-driver passes a Selenese command to the server.
  4. The Server interprets the command and then triggers the corresponding JavaScript execution to execute that command within the browser. Selenium-Core instructs the browser to act on that first instruction, typically opening a page of the AUT.
  5. The browser receives the open request and asks for the website’s content from the Selenium RC server (set as the HTTP proxy for the browser to use).
  6. Selenium RC server communicates with the Web server asking for the page and once it receives it, it sends the page to the browser masking the origin to look like the page comes from the same server as Selenium-Core (this allows Selenium-Core to comply with the Same Origin Policy).
  7. The browser receives the web page and renders it in the frame/window reserved for it.

Heightened Privileges Browsers

This workflow in this method is very similar to Proxy Injection but the main difference is that the browsers are launched in a special mode called Heightened Privileges, which allows websites to do things that are not commonly permitted (as doing XSS_, or filling file upload inputs and pretty useful stuff for Selenium). By using these browser modes, Selenium Core is able to directly open the AUT and read/interact with its content without having to pass the whole AUT through the Selenium RC server.

Here is the architectural diagram.

Architectural Diagram 1

As a test suite starts in your favorite language, the following happens:

  1. The client/driver establishes a connection with the selenium-RC server.
  2. Selenium RC server launches a browser (or reuses an old one) with a URL that will load Selenium-Core in the web page.
  3. Selenium-Core gets the first instruction from the client/driver (via another HTTP request made to the Selenium RC Server).
  4. Selenium-Core acts on that first instruction, typically opening a page of the AUT.
  5. The browser receives the open request and asks the Web Server for the page. Once the browser receives the web page, renders it in the frame/window reserved for it.

Handling HTTPS and Security Popups

Many applications switch from using HTTP to HTTPS when they need to send encrypted information such as passwords or credit card information. This is common with many of today’s web applications. Selenium RC supports this.

To ensure the HTTPS site is genuine, the browser will need a security certificate. Otherwise, when the browser accesses the AUT using HTTPS, it will assume that application is not ‘trusted’. When this occurs the browser displays security popups, and these popups cannot be closed using Selenium RC.

When dealing with HTTPS in a Selenium RC test, you must use a run mode that supports this and handles the security certificate for you. You specify the run mode when your test program initializes Selenium.

In Selenium RC 1.0 beta 2 and later use *firefox or *iexplore for the run mode. In earlier versions, including Selenium RC 1.0 beta 1, use *chrome or *iehta, for the run mode. Using these run modes, you will not need to install any special security certificates; Selenium RC will handle it for you.

In version 1.0 the run modes *firefox or *iexplore are recommended. However, there are additional run modes of *iexploreproxy and *firefoxproxy. These are provided for backwards compatibility only, and should not be used unless required by legacy test programs. Their use will present limitations with security certificate handling and with the running of multiple windows if your application opens additional browser windows.

In earlier versions of Selenium RC, *chrome or *iehta were the run modes that supported HTTPS and the handling of security popups. These were considered ‘experimental modes although they became quite stable and many people used them. If you are using Selenium 1.0 you do not need, and should not use, these older run modes.

Security Certificates Explained

Normally, your browser will trust the application you are testing by installing a security certificate which you already own. You can check this in your browser’s options or Internet properties (if you don’t know your AUT’s security certificate ask your system administrator). When Selenium loads your browser it injects code to intercept messages between the browser and the server. The browser now thinks untrusted software is trying to look like your application. It responds by alerting you with popup messages.

To get around this, Selenium RC, (again when using a run mode that support this) will install its own security certificate, temporarily, to your client machine in a place where the browser can access it. This tricks the browser into thinking it’s accessing a site different from your AUT and effectively suppresses the popups.

Another method used with earlier versions of Selenium was to install the Cybervillians security certificate provided with your Selenium installation. Most users should no longer need to do this however; if you are running Selenium RC in proxy injection mode, you may need to explicitly install this security certificate.

Supporting Additional Browsers and Browser Configurations

The Selenium API supports running against multiple browsers in addition to Internet Explorer and Mozilla Firefox. See the https://selenium.dev website for supported browsers. In addition, when a browser is not directly supported, you may still run your Selenium tests against a browser of your choosing by using the “*custom” run-mode (i.e. in place of *firefox or *iexplore) when your test application starts the browser. With this, you pass in the path to the browsers executable within the API call. This can also be done from the Server in interactive mode.

   cmd=getNewBrowserSession&1=*custom c:\Program Files\Mozilla Firefox\MyBrowser.exe&2=http://www.google.com

Running Tests with Different Browser Configurations

Normally Selenium RC automatically configures the browser, but if you launch the browser using the “*custom” run mode, you can force Selenium RC to launch the browser as-is, without using an automatic configuration.

For example, you can launch Firefox with a custom configuration like this:

   cmd=getNewBrowserSession&1=*custom c:\Program Files\Mozilla Firefox\firefox.exe&2=http://www.google.com

Note that when launching the browser this way, you must manually configure the browser to use the Selenium Server as a proxy. Normally this just means opening your browser preferences and specifying “localhost:4444” as an HTTP proxy, but instructions for this can differ radically from browser to browser. Consult your browser’s documentation for details.

Be aware that Mozilla browsers can vary in how they start and stop. One may need to set the MOZ_NO_REMOTE environment variable to make Mozilla browsers behave a little more predictably. Unix users should avoid launching the browser using a shell script; it’s generally better to use the binary executable (e.g. firefox-bin) directly.

Troubleshooting Common Problems

When getting started with Selenium RC there’s a few potential problems that are commonly encountered. We present them along with their solutions here.

Unable to Connect to Server

When your test program cannot connect to the Selenium Server, Selenium throws an exception in your test program. It should display this message or a similar one:

    "Unable to connect to remote server (Inner Exception Message: 
	No connection could be made because the target machine actively 
	refused it )"
    
	(using .NET and XP Service Pack 2) 

If you see a message like this, be sure you started the Selenium Server. If so, then there is a problem with the connectivity between the Selenium Client Library and the Selenium Server.

When starting with Selenium RC, most people begin by running their test program (with a Selenium Client Library) and the Selenium Server on the same machine. To do this use “localhost” as your connection parameter. We recommend beginning this way since it reduces the influence of potential networking problems which you’re getting started. Assuming your operating system has typical networking and TCP/IP settings you should have little difficulty. In truth, many people choose to run the tests this way.

If, however, you do want to run Selenium Server on a remote machine, the connectivity should be fine assuming you have valid TCP/IP connectivity between the two machines.

If you have difficulty connecting, you can use common networking tools like ping, telnet, ifconfig(Unix)/ipconfig (Windows), etc to ensure you have a valid network connection. If unfamilar with these, your system administrator can assist you.

Unable to Load the Browser

Ok, not a friendly error message, sorry, but if the Selenium Server cannot load the browser you will likely see this error.

    (500) Internal Server Error

This could be caused by

  • Firefox (prior to Selenium 1.0) cannot start because the browser is already open and you did not specify a separate profile. See the section on Firefox profiles under Server Options.
  • The run mode you’re using doesn’t match any browser on your machine. Check the parameters you passed to Selenium when you program opens the browser.
  • You specified the path to the browser explicitly (using “*custom”–see above) but the path is incorrect. Check to be sure the path is correct. Also check the user group to be sure there are no known issues with your browser and the “*custom” parameters.

Selenium Cannot Find the AUT

If your test program starts the browser successfully, but the browser doesn’t display the website you’re testing, the most likely cause is your test program is not using the correct URL.

This can easily happen. When you use Selenium-IDE to export your script, it inserts a dummy URL. You must manually change the URL to the correct one for your application to be tested.

Firefox Refused Shutdown While Preparing a Profile

This most often occurs when you run your Selenium RC test program against Firefox, but you already have a Firefox browser session running and, you didn’t specify a separate profile when you started the Selenium Server. The error from the test program looks like this:

    Error:  java.lang.RuntimeException: Firefox refused shutdown while 
    preparing a profile 

Here’s the complete error message from the server:

    16:20:03.919 INFO - Preparing Firefox profile... 
    16:20:27.822 WARN - GET /selenium-server/driver/?cmd=getNewBrowserSession&1=*fir 
    efox&2=http%3a%2f%2fsage-webapp1.qa.idc.com HTTP/1.1 
    java.lang.RuntimeException: Firefox refused shutdown while preparing a profile 
            at org.openqa.selenium.server.browserlaunchers.FirefoxCustomProfileLaunc 
    her.waitForFullProfileToBeCreated(FirefoxCustomProfileLauncher.java:277) 
    ... 
    Caused by: org.openqa.selenium.server.browserlaunchers.FirefoxCustomProfileLaunc 
    her$FileLockRemainedException: Lock file still present! C:\DOCUME~1\jsvec\LOCALS 
    ~1\Temp\customProfileDir203138\parent.lock 

To resolve this, see the section on Specifying a Separate Firefox Profile

Versioning Problems

Make sure your version of Selenium supports the version of your browser. For example, Selenium RC 0.92 does not support Firefox 3. At times you may be lucky (I was). But don’t forget to check which browser versions are supported by the version of Selenium you are using. When in doubt, use the latest release version of Selenium with the most widely used version of your browser.

Error message: “(Unsupported major.minor version 49.0)” while starting server

This error says you’re not using a correct version of Java. The Selenium Server requires Java 1.5 or higher.

To check double-check your java version, run this from the command line.

   java -version

You should see a message showing the Java version.

   java version "1.5.0_07"
   Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_07-b03)
   Java HotSpot(TM) Client VM (build 1.5.0_07-b03, mixed mode)

If you see a lower version number, you may need to update the JRE, or you may simply need to add it to your PATH environment variable.

404 error when running the getNewBrowserSession command

If you’re getting a 404 error while attempting to open a page on “http://www.google.com/selenium-server/", then it must be because the Selenium Server was not correctly configured as a proxy. The “selenium-server” directory doesn’t exist on google.com; it only appears to exist when the proxy is properly configured. Proxy Configuration highly depends on how the browser is launched with firefox, iexplore, opera, or custom.

  • iexplore: If the browser is launched using *iexplore, you could be having a problem with Internet Explorer’s proxy settings. Selenium Server attempts To configure the global proxy settings in the Internet Options Control Panel. You must make sure that those are correctly configured when Selenium Server launches the browser. Try looking at your Internet Options control panel. Click on the “Connections” tab and click on “LAN Settings”.

    • If you need to use a proxy to access the application you want to test, you’ll need to start Selenium Server with “-Dhttp.proxyHost”; see the Proxy Configuration_ for more details.
    • You may also try configuring your proxy manually and then launching the browser with *custom, or with *iehta browser launcher.
  • custom: When using *custom you must configure the proxy correctly(manually), otherwise you’ll get a 404 error. Double-check that you’ve configured your proxy settings correctly. To check whether you’ve configured the proxy correctly is to attempt to intentionally configure the browser incorrectly. Try configuring the browser to use the wrong proxy server hostname, or the wrong port. If you had successfully configured the browser’s proxy settings incorrectly, then the browser will be unable to connect to the Internet, which is one way to make sure that one is adjusting the relevant settings.

  • For other browsers (*firefox, *opera) we automatically hard-code the proxy for you, and so there are no known issues with this functionality. If you’re encountering 404 errors and have followed this user guide carefully post your results to user group for some help from the user community.

Permission Denied Error

The most common reason for this error is that your session is attempting to violate the same-origin policy by crossing domain boundaries (e.g., accesses a page from http://domain1 and then accesses a page from http://domain2) or switching protocols (moving from http://domainX to https://domainX).

This error can also occur when JavaScript attempts to find UI objects which are not yet available (before the page has completely loaded), or are no longer available (after the page has started to be unloaded). This is most typically encountered with AJAX pages which are working with sections of a page or subframes that load and/or reload independently of the larger page.

This error can be intermittent. Often it is impossible to reproduce the problem with a debugger because the trouble stems from race conditions which are not reproducible when the debugger’s overhead is added to the system. Permission issues are covered in some detail in the tutorial. Read the section about the The Same Origin Policy, Proxy Injection carefully.

Handling Browser Popup Windows

There are several kinds of “Popups” that you can get during a Selenium test. You may not be able to close these popups by running Selenium commands if they are initiated by the browser and not your AUT. You may need to know how to manage these. Each type of popup needs to be addressed differently.

  • HTTP basic authentication dialogs: These dialogs prompt for a username/password to login to the site. To login to a site that requires HTTP basic authentication, use a username and password in the URL, as described in RFC 1738_, like this: open(“http://myusername:myuserpassword@myexample.com/blah/blah/blah").

  • SSL certificate warnings: Selenium RC automatically attempts to spoof SSL certificates when it is enabled as a proxy; see more on this in the section on HTTPS. If your browser is configured correctly, you should never see SSL certificate warnings, but you may need to configure your browser to trust our dangerous “CyberVillains” SSL certificate authority. Again, refer to the HTTPS section for how to do this.

  • modal JavaScript alert/confirmation/prompt dialogs: Selenium tries to conceal those dialogs from you (by replacing window.alert, window.confirm and window.prompt) so they won’t stop the execution of your page. If you’re seeing an alert pop-up, it’s probably because it fired during the page load process, which is usually too early for us to protect the page. Selenese contains commands for asserting or verifying alert and confirmation popups. See the sections on these topics in Chapter 4.

On Linux, why isn’t my Firefox browser session closing?

On Unix/Linux you must invoke “firefox-bin” directly, so make sure that executable is on the path. If executing Firefox through a shell script, when it comes time to kill the browser Selenium RC will kill the shell script, leaving the browser running. You can specify the path to firefox-bin directly, like this.

   cmd=getNewBrowserSession&1=*firefox /usr/local/firefox/firefox-bin&2=http://www.google.com

Firefox *chrome doesn’t work with custom profile

Check Firefox profile folder -> prefs.js -> user_pref(“browser.startup.page”, 0); Comment this line like this: “//user_pref(“browser.startup.page”, 0);” and try again.

Is it ok to load a custom pop-up as the parent page is loading (i.e., before the parent page’s javascript window.onload() function runs)?

No. Selenium relies on interceptors to determine window names as they are being loaded. These interceptors work best in catching new windows if the windows are loaded AFTER the onload() function. Selenium may not recognize windows loaded before the onload function.

Firefox on Linux

On Unix/Linux, versions of Selenium before 1.0 needed to invoke “firefox-bin” directly, so if you are using a previous version, make sure that the real executable is on the path.

On most Linux distributions, the real firefox-bin is located on:

   /usr/lib/firefox-x.x.x/ 

Where the x.x.x is the version number you currently have. So, to add that path to the user’s path. you will have to add the following to your .bashrc file:

   export PATH="$PATH:/usr/lib/firefox-x.x.x/"

If necessary, you can specify the path to firefox-bin directly in your test, like this:

   "*firefox /usr/lib/firefox-x.x.x/firefox-bin"

IE and Style Attributes

If you are running your tests on Internet Explorer and you cannot locate elements using their style attribute. For example:

    //td[@style="background-color:yellow"]

This would work perfectly in Firefox, Opera or Safari but not with IE. IE interprets the keys in @style as uppercase. So, even if the source code is in lowercase, you should use:

    //td[@style="BACKGROUND-COLOR:yellow"]

This is a problem if your test is intended to work on multiple browsers, but you can easily code your test to detect the situation and try the alternative locator that only works in IE.

Error encountered - “Cannot convert object to primitive value” with shut down of *googlechrome browser

To avoid this error you have to start browser with an option that disables same origin policy checks:

   selenium.start("commandLineFlags=--disable-web-security");

Error encountered in IE - “Couldn’t open app window; is the pop-up blocker enabled?”

To avoid this error you have to configure the browser: disable the popup blocker AND uncheck ‘Enable Protected Mode’ option in Tools » Options » Security.


  1. The proxy is a third person in the middle that passes the ball between the two parts. It acts as a “web server” that delivers the AUT to the browser. Being a proxy gives Selenium Server the capability of “lying” about the AUT’s real URL. ↩︎

  2. The browser is launched with a configuration profile that has set localhost:4444 as the HTTP proxy, this is why any HTTP request that the browser does will pass through Selenium server and the response will pass through it and not from the real server. ↩︎

2 - Migrating from RC to WebDriver

Comment migrer vers Selenium WebDriver

Une question fréquente lors de l’adoption de Selenium 2 est la bonne chose à faire lors de l’ajout de nouveaux tests à un ensemble de tests existant? Utilisateurs nouveaux dans le Le framework peut commencer par utiliser les nouvelles API WebDriver pour écrire leurs tests. Mais qu’en est-il des utilisateurs qui ont déjà des suites de tests existants? Ce guide est conçu pour montrer comment migrer vos tests existants vers les nouvelles API, permettant à tous les nouveaux tests d’être écrits en utilisant les nouvelles fonctionnalités offertes par WebDriver.

La méthode présentée ici décrit une migration fragmentaire vers le WebDriver API sans avoir à tout retravailler en une seule poussée massive. Ça signifie que vous pouvez accorder plus de temps pour la migration de vos tests existants, qui peut vous aider à décider où dépenser vos efforts.

Ce guide est écrit en utilisant Java, car il a le meilleur support pour faire la migration. Comme nous fournissons de meilleurs outils pour d’autres langues, ce guide sera élargi pour inclure ces langues.

Pourquoi migrer vers WebDriver

Le déplacement d’une suite de tests d’une API à une autre API nécessite une énorme quantité d’effort. Pourquoi voudriez-vous, vous et votre équipe, envisager cette démarche? Voici quelques raisons pour lesquelles vous devriez envisager de migrer vos tests de Selenium pour utiliser WebDriver.

  • API plus petite et compacte. L’API de WebDriver est plus orientée objet que le API d’origine Selenium RC. Cela peut faciliter le travail avec.
  • Meilleure émulation des interactions utilisateur. Dans la mesure du possible, WebDriver fait utilisation d’événements natifs pour interagir avec une page Web. Cela de plus près imite la façon dont vos utilisateurs travaillent avec votre site et vos applications. En plus, WebDriver propose les API d’interaction utilisateur avancées qui vous permettent de modélisez des interactions complexes avec votre site.
  • Prise en charge par les fournisseurs de navigateurs. Opera, Mozilla et Google sont tous actifs participants au développement de WebDriver, et chacun a des ingénieurs travaillant pour améliorer le cadre. Souvent, cela signifie que la prise en charge de WebDriver est intégré dans le navigateur lui-même: vos tests s’exécutent aussi rapidement et aussi stable que possible.

Avant de commencer

Afin de rendre le processus de migration aussi indolore que possible, faites assurez-vous que tous vos tests fonctionnent correctement avec la dernière version de Selenium. Cela peut sembler évident, mais il vaut mieux le dire!

Commencer

La première étape lors du démarrage de la migration consiste à modifier la façon dont vous obtenez votre instance de Selenium. Lorsque vous utilisez Selenium RC, cela se fait comme suit:

Selenium selenium = new DefaultSelenium("localhost", 4444, "*firefox", "http://www.yoursite.com");
selenium.start();

Cela devrait être remplacé comme suit:

WebDriver driver = new FirefoxDriver();
Selenium selenium = new WebDriverBackedSelenium(driver, "http://www.yoursite.com");

Prochaines étapes

Une fois vos tests exécutés sans erreur, la prochaine étape consiste à migrer le code de test réel pour utiliser les API WebDriver. Selon la façon dont bien abstrait votre code est, cela peut être un processus court ou long. Dans les deux cas, l’approche est la même et peut se résumer simplement: modifier le code pour utiliser la nouvelle API lorsque vous venez de le modifier.

Si vous devez extraire l’implémentation WebDriver sous-jacente de l’instance de Selenium, vous pouvez simplement la lancer sur WrapsDriver:

WebDriver driver = ((WrapsDriver) selenium).getWrappedDriver();

Cela vous permet de continuer à transmettre l’instance Selenium comme normal, mais pour déballer l’instance WebDriver comme requis.

À un moment donné, votre base de code utilisera principalement les nouvelles API. À ce stade, vous pouvez inverser la relation en utilisant WebDriver et instancier une instance de Selenium à la demande:

Selenium selenium = new WebDriverBackedSelenium(driver, baseUrl);

Problèmes communs

Heureusement, vous n’êtes pas la première personne à traverser cette migration, voici donc quelques problèmes communs que d’autres ont vus, et comment les résoudre.

Cliquer et taper est plus complet

Un schéma courant dans un test Selenium RC est de voir quelque chose comme:

selenium.type("name", "exciting tex");
selenium.keyDown("name", "t");
selenium.keyPress("name", "t");
selenium.keyUp("name", "t");

Cela repose sur le fait que “type” remplace simplement le contenu du élément identifié sans déclencher également tous les événements qui normalement être renvoyé si un utilisateur interagit avec la page. Les invocations directes finales de “clé*” provoque le déclenchement des gestionnaires JS comme prévu.

Lors de l’utilisation de WebDriverBackedSelenium, le résultat du remplissage du formulaire serait “texttt excitant”: pas ce que vous attendez! La raison de cela est que WebDriver émule plus précisément le comportement des utilisateurs, et aura donc a tiré des événements tout au long.

Ce même fait peut parfois déclencher un chargement de page plus tôt qu’il ne le ferait faire dans un test de sélénium 1. Vous pouvez dire que cela s’est produit si un “StaleElementException” est levé par WebDriver.

WaitForPageToLoad revient trop tôt

Découvrir quand un chargement de page est terminé est une entreprise délicate. Voulons-nous “lorsque l’événement de chargement se déclenche”, “lorsque toutes les requêtes AJAX sont terminées”, “lorsque il n’y a pas de trafic réseau “,” lorsque document.readyState a changé “ou quelque chose sinon entièrement?

WebDriver tente de simuler le comportement d’origine de Selenium, mais cela ne fonctionnent toujours parfaitement pour diverses raisons. La raison la plus courante est que c’est difficile de faire la différence entre un chargement de page n’ayant pas encore commencé et un chargement de page terminé entre les appels de méthode. Cela signifie parfois que le contrôle revient à votre test avant que la page ne soit terminée (ou même commencée!) chargement.

La solution est d’attendre quelque chose de spécifique. Généralement, cela pourrait être pour l’élément avec lequel vous souhaitez interagir ensuite, ou pour une variable Javascript à définir sur une valeur spécifique. Un exemple serait:

Wait<WebDriver> wait = new WebDriverWait(driver, Duration.ofSeconds(30));
WebElement element= wait.until(visibilityOfElementLocated(By.id("some_id")));

Où “visibilityOfElementLocated” est implémenté comme:

public ExpectedCondition<WebElement> visibilityOfElementLocated(final By locator) {
  return new ExpectedCondition<WebElement>() {
    public WebElement apply(WebDriver driver) {
      WebElement toReturn = driver.findElement(locator);
      if (toReturn.isDisplayed()) {
        return toReturn;
      }
      return null;
    }
  };
}

Cela peut sembler complexe, mais c’est presque tout le code de la plaque de la chaudière. Le seul bit intéressant est que le “ExpectedCondition” sera évalué à plusieurs reprises jusqu’à ce que la méthode “apply” renvoie quelque chose qui n’est ni “null” ni Boolean.FALSE.

Bien sûr, l’ajout de tous ces appels “d’attente” peut encombrer votre code. Si c’est le cas, et vos besoins sont simples, pensez à utiliser les attentes implicites:

driver.manage().timeouts().implicitlyWait(30, TimeUnit.SECONDS);

En faisant cela, chaque fois qu’un élément est localisé, si l’élément n’est pas présent, l’emplacement est réessayé jusqu’à ce qu’il soit présent ou jusqu’à 30 secondes passé.

Trouver par XPath ou sélecteurs CSS ne fonctionne pas toujours, mais cela fonctionne avec Selenium 1

Dans Selenium 1, il était courant que xpath utilise une bibliothèque groupée plutôt que les capacités du navigateur lui-même. WebDriver utilisera toujours le natif méthodes de navigateur, sauf s’il n’y a pas d’autre alternative. Cela signifie que xpath complexe les expressions peuvent se briser sur certains navigateurs.

Les sélecteurs CSS dans Selenium 1 ont été implémentés à l’aide de la bibliothèque Sizzle. Cette implémente un surensemble de la spécification CSS Selector, et il n’est pas toujours clair où vous avez franchi la ligne. Si vous utilisez WebDriverBackedSelenium et utilisez un Sizzle locator au lieu d’un CSS Selector pour trouver des éléments, un avertissement être connecté à la console. Cela vaut la peine de prendre le temps de les chercher, en particulier si les tests échouent faute de pouvoir trouver des éléments.

Il n’y a pas de Browserbot

Selenium RC était basé sur Selenium Core, et donc lorsque vous avez exécuté JavaScript, vous pouvez accéder à des morceaux de Selenium Core pour faciliter les choses. Comme WebDriver n’est pas basé sur Selenium Core, cela n’est plus possible. Comment savoir si vous utilisez Selenium Core? Facile! Regardez juste pour voir si vos appels “getEval” ou similaires utilisent “Selenium” ou “browserbot” dans le Javascript évalué.

Vous utilisez peut-être le navigateur pour obtenir un descripteur de la fenêtre actuelle ou document du test. Heureusement, WebDriver évalue toujours JS dans le contexte de la fenêtre en cours, vous pouvez donc utiliser directement “fenêtre” ou “document”.

Vous pouvez également utiliser le navigateur pour localiser des éléments. Dans WebDriver, l’idiome pour ce faire est de localiser d’abord l’élément, puis passez cela comme argument au Javascript. Donc:

String name = selenium.getEval(
    "selenium.browserbot.findElement('id=foo', browserbot.getCurrentWindow()).tagName");

devient:

WebElement element = driver.findElement(By.id("foo"));
String name = (String) ((JavascriptExecutor) driver).executeScript(
    "return arguments[0].tagName", element);

Remarquez comment la variable “element” passée apparaît comme premier élément dans le tableau “arguments” standard de JS.

Executing Javascript Doesn’t Return Anything

JavascriptExecutor de WebDriver encapsulera tous les JS et l’évaluera comme une expression anonyme. Cela signifie que vous devez utiliser le mot-clé “retour”:

String title = selenium.getEval("browserbot.getCurrentWindow().document.title");

devient:

((JavascriptExecutor) driver).executeScript("return document.title;");

3 - Runner HTML

Selenium HTML-runner permet d’exécuter des Test Suites depuis la ligne de commande. Les Tests Suites sont des exports HTML depuis Selenium IDE ou autre outil compatible.

Informations générales

  • Les combinaisons de version de geckodriver / firefox / selenium-html-runner sont importantes. Il doit y avoir une matrice de compatibilité quelque part.
  • selenium-html-runner exécute seulement des Test Suites (et non pas des Test Case, comme un export de Monitis Transaction Monitor). Assurez-vous de vous y conformer.
  • Pour les utilisateurs Linux sans affichage - vous devez exécuter html-runner avec un affichage virtuel (cherchez xvfb)

Exemple sur environnement Linux

Installer / télécharger les packages suivants:

[user@localhost ~]$ cat /etc/redhat-release
CentOS Linux release 7.4.1708 (Core)

[user@localhost ~]$ rpm -qa | egrep -i "xvfb|java-1.8|firefox"
xorg-x11-server-Xvfb-1.19.3-11.el7.x86_64
firefox-52.4.0-1.el7.centos.x86_64
java-1.8.0-openjdk-1.8.0.151-1.b12.el7_4.x86_64
java-1.8.0-openjdk-headless-1.8.0.151-1.b12.el7_4.x86_64

Example de Test Suite:

[user@localhost ~]$ cat testsuite.html
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
  <meta content="text/html; charset=UTF-8" http-equiv="content-type" />
  <title>Test Suite</title>
</head>
<body>
<table id="suiteTable" cellpadding="1" cellspacing="1" border="1" class="selenium"><tbody>
<tr><td><b>Test Suite</b></td></tr>
<tr><td><a href="YOUR-TEST-SCENARIO.html">YOUR-TEST-SCENARIO</a></td></tr>
</tbody></table>
</body>
</html>

Comment exécuter selenium-html-runner en mode headless

Pour la partie la plus importante, voici un exemple d’exécution de selenium-html-runner ! Votre expérience peut varier selon les combinaisons logicielles - versions de geckodriver/FF/html-runner.

xvfb-run java -Dwebdriver.gecko.driver=/home/mmasek/geckodriver.0.18.0 -jar selenium-html-runner-3.7.1.jar -htmlSuite "firefox" "https://YOUR-BASE-URL" "$(pwd)/testsuite.html" "results.html" ; grep result: -A1 results.html/firefox.results.html
[user@localhost ~]$ xvfb-run java -Dwebdriver.gecko.driver=/home/mmasek/geckodriver.0.18.0 -jar selenium-html-runner-3.7.1.jar -htmlSuite "*firefox" "https://YOUR-BASE-URL" "$(pwd)/testsuite.html" "results.html" ; grep result: -A1 results.html/firefox.results.html
Multi-window mode is longer used as an option and will be ignored.
1510061109691   geckodriver     INFO    geckodriver 0.18.0
1510061109708   geckodriver     INFO    Listening on 127.0.0.1:2885
1510061110162   geckodriver::marionette INFO    Starting browser /usr/bin/firefox with args ["-marionette"]
1510061111084   Marionette      INFO    Listening on port 43229
1510061111187   Marionette      WARN    TLS certificate errors will be ignored for this session
Nov 07, 2017 1:25:12 PM org.openqa.selenium.remote.ProtocolHandshake createSession
INFO: Detected dialect: W3C
2017-11-07 13:25:12.714:INFO::main: Logging initialized @3915ms to org.seleniumhq.jetty9.util.log.StdErrLog
2017-11-07 13:25:12.804:INFO:osjs.Server:main: jetty-9.4.z-SNAPSHOT
2017-11-07 13:25:12.822:INFO:osjsh.ContextHandler:main: Started o.s.j.s.h.ContextHandler@87a85e1{/tests,null,AVAILABLE}
2017-11-07 13:25:12.843:INFO:osjs.AbstractConnector:main: Started ServerConnector@52102734{HTTP/1.1,[http/1.1]}{0.0.0.0:31892}
2017-11-07 13:25:12.843:INFO:osjs.Server:main: Started @4045ms
Nov 07, 2017 1:25:13 PM org.openqa.selenium.server.htmlrunner.CoreTestCase run
INFO: |open | /auth_mellon.php |  |
Nov 07, 2017 1:25:14 PM org.openqa.selenium.server.htmlrunner.CoreTestCase run
INFO: |waitForPageToLoad | 3000 |  |
.
.
.etc

<td>result:</td>
<td>PASS</td>


4 - Legacy Selenium IDE

Most of the documentation found in this section is still in English. Please note we are not accepting pull requests to translate this content as translating documentation of legacy components does not add value to the community nor the project.

Introduction

The Selenium-IDE (Integrated Development Environment) is the tool you use to develop your Selenium test cases. It’s an easy-to-use Firefox plug-in and is generally the most efficient way to develop test cases. It also contains a context menu that allows you to first select a UI element from the browser’s currently displayed page and then select from a list of Selenium commands with parameters pre-defined according to the context of the selected UI element. This is not only a time-saver, but also an excellent way of learning Selenium script syntax.

This chapter is all about the Selenium IDE and how to use it effectively.

Installing the IDE

Using Firefox, first, download the IDE from the SeleniumHQ downloads page

Firefox will protect you from installing addons from unfamiliar locations, so you will need to click ‘Allow’ to proceed with the installation, as shown in the following screenshot.

Selenium IDE Installation 1

When downloading from Firefox, you’ll be presented with the following window.

Selenium IDE Installation 2

Select Install Now. The Firefox Add-ons window pops up, first showing a progress bar, and when the download is complete, displays the following.

Selenium IDE Installation 3

Restart Firefox. After Firefox reboots you will find the Selenium-IDE listed under the Firefox Tools menu.

Selenium IDE Installation 4

Opening the IDE

To run the Selenium-IDE, simply select it from the Firefox Tools menu. It opens as follows with an empty script-editing window and a menu for loading, or creating new test cases.

Selenium IDE Open

IDE Features

The File menu has options for Test Case and Test Suite (suite of Test Cases). Using these you can add a new Test Case, open a Test Case, save a Test Case, export Test Case in a language of your choice. You can also open the recent Test Case. All these options are also available for Test Suite.

The Edit menu allows copy, paste, delete, undo, and select all operations for editing the commands in your test case. The Options menu allows the changing of settings. You can set the timeout value for certain commands, add user-defined user extensions to the base set of Selenium commands, and specify the format (language) used when saving your test cases. The Help menu is the standard Firefox Help menu; only one item on this menu–UI-Element Documentation–pertains to Selenium-IDE.

Toolbar

The toolbar contains buttons for controlling the execution of your test cases, including a step feature for debugging your test cases. The right-most button, the one with the red-dot, is the record button.

Selenium IDE Features Selenium IDE Features

Speed Control: controls how fast your test case runs.

Selenium IDE Features

Run All: Runs the entire test suite when a test suite with multiple test cases is loaded.

Selenium IDE Features

Run: Runs the currently selected test. When only a single test is loaded this button and the Run All button have the same effect.

Selenium IDE Features Selenium IDE Features

Pause/Resume: Allows stopping and re-starting of a running test case.

Selenium IDE Features

Step: Allows you to “step” through a test case by running it one command at a time. Use for debugging test cases.

Selenium IDE Features

TestRunner Mode: Allows you to run the test case in a browser loaded with the Selenium-Core TestRunner. The TestRunner is not commonly used now and is likely to be deprecated. This button is for evaluating test cases for backwards compatibility with the TestRunner. Most users will probably not need this button.

Selenium IDE Features

Apply Rollup Rules: This advanced feature allows repetitive sequences of Selenium commands to be grouped into a single action. Detailed documentation on rollup rules can be found in the UI-Element Documentation on the Help menu.

Selenium IDE Features

Test Case Pane

Your script is displayed in the test case pane. It has two tabs, one for displaying the command and their parameters in a readable “table” format.

Selenium IDE Image Pane

The other tab - Source displays the test case in the native format in which the file will be stored. By default, this is HTML although it can be changed to a programming language such as Java or C#, or a scripting language like Python. See the Options menu for details. The Source view also allows one to edit the test case in its raw form, including copy, cut and paste operations.

The Command, Target, and Value entry fields display the currently selected command along with its parameters. These are entry fields where you can modify the currently selected command. The first parameter specified for a command in the Reference tab of the bottom pane always goes in the Target field. If a second parameter is specified by the Reference tab, it always goes in the Value field.

Selenium IDE Entry Fields

If you start typing in the Command field, a drop-down list will be populated based on the first characters you type; you can then select your desired command from the drop-down.

Log/Reference/UI-Element/Rollup Pane

The bottom pane is used for four different functions–Log, Reference, UI-Element, and Rollup–depending on which tab is selected.

Log

When you run your test case, error messages and information messages showing the progress are displayed in this pane automatically, even if you do not first select the Log tab. These messages are often useful for test case debugging. Notice the Clear button for clearing the Log. Also notice the Info button is a drop-down allowing selection of different levels of information to log.

Selenium IDE Bottom Box

Reference

The Reference tab is the default selection whenever you are entering or modifying Selenese commands and parameters in Table mode. In Table mode, the Reference pane will display documentation on the current command. When entering or modifying commands, whether from Table or Source mode, it is critically important to ensure that the parameters specified in the Target and Value fields match those specified in the parameter list in the Reference pane. The number of parameters provided must match the number specified, the order of parameters provided must match the order specified, and the type of parameters provided must match the type specified. If there is a mismatch in any of these three areas, the command will not run correctly.

Selenium IDE Bottom Box

While the Reference tab is invaluable as a quick reference, it is still often necessary to consult the Selenium Reference document.

UI-Element and Rollup

Detailed information on these two panes (which cover advanced features) can be found in the UI-Element Documentation on the Help menu of Selenium-IDE.

Building Test Cases

There are three primary methods for developing test cases. Frequently, a test developer will require all three techniques.

Recording

Many first-time users begin by recording a test case from their interactions with a website. When Selenium-IDE is first opened, the record button is ON by default. If you do not want Selenium-IDE to begin recording automatically you can turn this off by going under Options > Options… and deselecting “Start recording immediately on open.”

During recording, Selenium-IDE will automatically insert commands into your test case based on your actions. Typically, this will include:

  • clicking a link - click or clickAndWait commands
  • entering values - type command
  • selecting options from a drop-down listbox - select command
  • clicking checkboxes or radio buttons - click command

Here are some “gotchas” to be aware of:

  • The type command may require clicking on some other area of the web page for it to record.
  • Following a link usually records a click command. You will often need to change this to clickAndWait to ensure your test case pauses until the new page is completely loaded. Otherwise, your test case will continue running commands before the page has loaded all its UI elements. This will cause unexpected test case failures.

Adding Verifications and Asserts With the Context Menu

Your test cases will also need to check the properties of a web-page. This requires assert and verify commands. We won’t describe the specifics of these commands here; that is in the chapter on Selenium Commands – “Selenese”. Here we’ll simply describe how to add them to your test case.

With Selenium-IDE recording, go to the browser displaying your test application and right click anywhere on the page. You will see a context menu showing verify and/or assert commands.

The first time you use Selenium, there may only be one Selenium command listed. As you use the IDE however, you will find additional commands will quickly be added to this menu. Selenium-IDE will attempt to predict what command, along with the parameters, you will need for a selected UI element on the current web-page.

Let’s see how this works. Open a web-page of your choosing and select a block of text on the page. A paragraph or a heading will work fine. Now, right-click the selected text. The context menu should give you a verifyTextPresent command and the suggested parameter should be the text itself.

Also, notice the Show All Available Commands menu option. This shows many, many more commands, again, along with suggested parameters, for testing your currently selected UI element.

Try a few more UI elements. Try right-clicking an image, or a user control like a button or a checkbox. You may need to use Show All Available Commands to see options other than verifyTextPresent. Once you select these other options, the more commonly used ones will show up on the primary context menu. For example, selecting verifyElementPresent for an image should later cause that command to be available on the primary context menu the next time you select an image and right-click.

Again, these commands will be explained in detail in the chapter on Selenium commands. For now though, feel free to use the IDE to record and select commands into a test case and then run it. You can learn a lot about the Selenium commands simply by experimenting with the IDE.

Editing

Insert Command

Table View

Select the point in your test case where you want to insert the command. To do this, in the Test Case Pane, left-click on the line where you want to insert a new command. Right-click and select Insert Command; the IDE will add a blank line just ahead of the line you selected. Now use the command editing text fields to enter your new command and its parameters.

Source View

Select the point in your test case where you want to insert the command. To do this, in the Test Case Pane, left-click between the commands where you want to insert a new command, and enter the HTML tags needed to create a 3-column row containing the Command, first parameter (if one is required by the Command), and second parameter (again, if one is required to locate an element) and third parameter(again, if one is required to have a value). Example:

<tr>
    <td>Command</td>
    <td>target (locator)</td>
    <td>Value</td>
</tr>

Insert Comment

Comments may be added to make your test case more readable. These comments are ignored when the test case is run.

Comments may also be used to add vertical white space (one or more blank lines) in your tests; just create empty comments. An empty command will cause an error during execution; an empty comment won’t.

Table View

Select the line in your test case where you want to insert the comment. Right-click and select Insert Comment. Now use the Command field to enter the comment. Your comment will appear in purple text.

Source View

Select the point in your test case where you want to insert the comment. Add an HTML-style comment, i.e., <!-- your comment here -->.

Edit a Command or Comment

Table View

Simply select the line to be changed and edit it using the Command, Target, and Value fields.

Source View

Since Source view provides the equivalent of a WYSIWYG (What You See is What You Get) editor, simply modify which line you wish–command, parameter, or comment.

Opening and Saving a Test Case

Like most programs, there are Save and Open commands under the File menu. However, Selenium distinguishes between test cases and test suites. To save your Selenium-IDE tests for later use you can either save the individual test cases, or save the test suite. If the test cases of your test suite have not been saved, you’ll be prompted to save them before saving the test suite.

When you open an existing test case or suite, Selenium-IDE displays its Selenium commands in the Test Case Pane.

Running Test Cases

The IDE allows many options for running your test case. You can run a test case all at once, stop and start it, run it one line at a time, run a single command you are currently developing, and you can do a batch run of an entire test suite. Execution of test cases is very flexible in the IDE.

Run a Test Case

Click the Run button to run the currently displayed test case.

Run a Test Suite

Click the Run All button to run all the test cases in the currently loaded test suite.

Stop and Start

The Pause button can be used to stop the test case while it is running. The icon of this button then changes to indicate the Resume button. To continue click Resume.

Stop in the Middle

You can set a breakpoint in the test case to cause it to stop on a particular command. This is useful for debugging your test case. To set a breakpoint, select a command, right-click, and from the context menu select Toggle Breakpoint.

Start from the Middle

You can tell the IDE to begin running from a specific command in the middle of the test case. This also is used for debugging. To set a startpoint, select a command, right-click, and from the context menu select Set/Clear Start Point.

Run Any Single Command

Double-click any single command to run it by itself. This is useful when writing a single command. It lets you immediately test a command you are constructing, when you are not sure if it is correct. You can double-click it to see if it runs correctly. This is also available from the context menu.

Using Base URL to Run Test Cases in Different Domains

The Base URL field at the top of the Selenium-IDE window is very useful for allowing test cases to be run across different domains. Suppose that a site named http://news.portal.com had an in-house beta site named http://beta.news.portal.com. Any test cases for these sites that begin with an open statement should specify a relative URL as the argument to open rather than an absolute URL (one starting with a protocol such as http: or https:). Selenium-IDE will then create an absolute URL by appending the open command’s argument onto the end of the value of Base URL. For example, the test case below would be run against http://news.portal.com/about.html:

Selenium IDE Prod URL

This same test case with a modified Base URL setting would be run against http://beta.news.portal.com/about.html:

Selenium IDE Beta URL

Selenium Commands – “Selenese”

Selenium commands, often called selenese, are the set of commands that run your tests. A sequence of these commands is a test script. Here we explain those commands in detail, and we present the many choices you have in testing your web application when using Selenium.

Selenium provides a rich set of commands for fully testing your web-app in virtually any way you can imagine. The command set is often called selenese. These commands essentially create a testing language.

In selenese, one can test the existence of UI elements based on their HTML tags, test for specific content, test for broken links, input fields, selection list options, submitting forms, and table data among other things. In addition Selenium commands support testing of window size, mouse position, alerts, Ajax functionality, pop up windows, event handling, and many other web-application features. The Command Reference lists all the available commands.

A command tells Selenium what to do. Selenium commands come in three “flavors”: Actions, Accessors, and Assertions.

  • Actions are commands that generally manipulate the state of the application. They do things like “click this link” and “select that option”. If an Action fails, or has an error, the execution of the current test is stopped.

    Many Actions can be called with the “AndWait” suffix, e.g. “clickAndWait”. This suffix tells Selenium that the action will cause the browser to make a call to the server, and that Selenium should wait for a new page to load.

  • Accessors examine the state of the application and store the results in variables, e.g. “storeTitle”. They are also used to automatically generate Assertions.

  • Assertions are like Accessors, but they verify that the state of the application conforms to what is expected. Examples include “make sure the page title is X” and “verify that this checkbox is checked”.

All Selenium Assertions can be used in 3 modes: “assert”, “verify”, and ” waitFor”. For example, you can “assertText”, “verifyText” and “waitForText”. When an “assert” fails, the test is aborted. When a “verify” fails, the test will continue execution, logging the failure. This allows a single “assert” to ensure that the application is on the correct page, followed by a bunch of “verify” assertions to test form field values, labels, etc.

“waitFor” commands wait for some condition to become true (which can be useful for testing Ajax applications). They will succeed immediately if the condition is already true. However, they will fail and halt the test if the condition does not become true within the current timeout setting (see the setTimeout action below).

Script Syntax

Selenium commands are simple, they consist of the command and two parameters. For example:

verifyText//div//a[2]Login

The parameters are not always required; it depends on the command. In some cases both are required, in others one parameter is required, and in still others the command may take no parameters at all. Here are a couple more examples:

goBackAndWait
verifyTextPresentWelcome to My Home Page
typeid=phone(555) 666-7066
typeid=address1${myVariableAddress}

The command reference describes the parameter requirements for each command.

Parameters vary, however they are typically:

  • a locator for identifying a UI element within a page.
  • a text pattern for verifying or asserting expected page content
  • a text pattern or a Selenium variable for entering text in an input field or for selecting an option from an option list.

Locators, text patterns, Selenium variables, and the commands themselves are described in considerable detail in the section on Selenium Commands.

Selenium scripts that will be run from Selenium-IDE will be stored in an HTML text file format. This consists of an HTML table with three columns. The first column identifies the Selenium command, the second is a target, and the final column contains a value. The second and third columns may not require values depending on the chosen Selenium command, but they should be present. Each table row represents a new Selenium command. Here is an example of a test that opens a page, asserts the page title and then verifies some content on the page:

<table>
    <tr><td>open</td><td>/download/</td><td></td></tr>
    <tr><td>assertTitle</td><td></td><td>Downloads</td></tr>
    <tr><td>verifyText</td><td>//h2</td><td>Downloads</td></tr>
</table>

Rendered as a table in a browser this would look like the following:

open/download/
assertTitleDownloads
verifyText//h2Downloads

The Selenese HTML syntax can be used to write and run tests without requiring knowledge of a programming language. With a basic knowledge of selenese and Selenium-IDE you can quickly produce and run testcases.

Test Suites

A test suite is a collection of tests. Often one will run all the tests in a test suite as one continuous batch-job.

When using Selenium-IDE, test suites also can be defined using a simple HTML file. The syntax again is simple. An HTML table defines a list of tests where each row defines the filesystem path to each test. An example tells it all.

<html>
<head>
<title>Test Suite Function Tests - Priority 1</title>
</head>
<body>
<table>
  <tr><td><b>Suite Of Tests</b></td></tr>
  <tr><td><a href="./Login.html">Login</a></td></tr>
  <tr><td><a href="./SearchValues.html">Test Searching for Values</a></td></tr>
  <tr><td><a href="./SaveValues.html">Test Save</a></td></tr>
</table>
</body>
</html>

A file similar to this would allow running the tests all at once, one after another, from the Selenium-IDE.

Test suites can also be maintained when using Selenium-RC. This is done via programming and can be done a number of ways. Commonly Junit is used to maintain a test suite if one is using Selenium-RC with Java. Additionally, if C# is the chosen language, Nunit could be employed. If using an interpreted language like Python with Selenium-RC then some simple programming would be involved in setting up a test suite. Since the whole reason for using Selenium-RC is to make use of programming logic for your testing this usually isn’t a problem.

Commonly Used Selenium Commands

To conclude our introduction of Selenium, we’ll show you a few typical Selenium commands. These are probably the most commonly used commands for building tests.

open

opens a page using a URL.

click/clickAndWait

performs a click operation, and optionally waits for a new page to load.

verifyTitle/assertTitle

verifies an expected page title.

verifyTextPresent

verifies expected text is somewhere on the page.

verifyElementPresent

verifies an expected UI element, as defined by its HTML tag, is present on the page.

verifyText

verifies expected text and its corresponding HTML tag are present on the page.

verifyTable

verifies a table’s expected contents.

waitForPageToLoad

pauses execution until an expected new page loads. Called automatically when clickAndWait is used.

waitForElementPresent

pauses execution until an expected UI element, as defined by its HTML tag, is present on the page.

Verifying Page Elements

Verifying UI elements on a web page is probably the most common feature of your automated tests. Selenese allows multiple ways of checking for UI elements. It is important that you understand these different methods because these methods define what you are actually testing.

For example, will you test that…

  1. an element is present somewhere on the page?
  2. specific text is somewhere on the page?
  3. specific text is at a specific location on the page?

For example, if you are testing a text heading, the text and its position at the top of the page are probably relevant for your test. If, however, you are testing for the existence of an image on the home page, and the web designers frequently change the specific image file along with its position on the page, then you only want to test that an image (as opposed to the specific image file) exists somewhere on the page.

Assertion or Verification?

Choosing between “assert” and “verify” comes down to convenience and management of failures. There’s very little point checking that the first paragraph on the page is the correct one if your test has already failed when checking that the browser is displaying the expected page. If you’re not on the correct page, you’ll probably want to abort your test case so that you can investigate the cause and fix the issue(s) promptly. On the other hand, you may want to check many attributes of a page without aborting the test case on the first failure as this will allow you to review all failures on the page and take the appropriate action. Effectively an “assert” will fail the test and abort the current test case, whereas a “verify” will fail the test and continue to run the test case.

The best use of this feature is to logically group your test commands, and start each group with an “assert” followed by one or more “verify” test commands. An example follows:

CommandTargetValue
open/download/
assertTitleDownloads
verifyText//h2Downloads
assertTable1.2.1Selenium IDE
verifyTable1.2.2June 3, 2008
verifyTable1.2.31.0 beta 2

The above example first opens a page and then “asserts” that the correct page is loaded by comparing the title with the expected value. Only if this passes will the following command run and “verify” that the text is present in the expected location. The test case then “asserts” the first column in the second row of the first table contains the expected value, and only if this passed will the remaining cells in that row be “verified”.

verifyTextPresent

The command verifyTextPresent is used to verify specific text exists somewhere on the page. It takes a single argument–the text pattern to be verified. For example:

CommandTargetValue
verifyTextPresentMarketing Analysis

This would cause Selenium to search for, and verify, that the text string “Marketing Analysis” appears somewhere on the page currently being tested. Use verifyTextPresent when you are interested in only the text itself being present on the page. Do not use this when you also need to test where the text occurs on the page.

verifyElementPresent

Use this command when you must test for the presence of a specific UI element, rather than its content. This verification does not check the text, only the HTML tag. One common use is to check for the presence of an image.

CommandTargetValue
verifyElementPresent//div/p/img

This command verifies that an image, specified by the existence of an HTML tag, is present on the page, and that it follows a

tag and a

tag. The first (and only) parameter is a locator for telling the Selenese command how to find the element. Locators are explained in the next section.

verifyElementPresent can be used to check the existence of any HTML tag within the page. You can check the existence of links, paragraphs, divisions

, etc. Here are a few more examples.

CommandTargetValue
verifyElementPresent//div/p
verifyElementPresent//div/a
verifyElementPresentid=Login
verifyElementPresentlink=Go to Marketing Research
verifyElementPresent//a[2]
verifyElementPresent//head/title

These examples illustrate the variety of ways a UI element may be tested. Again, locators are explained in the next section.

verifyText

Use verifyText when both the text and its UI element must be tested. verifyText must use a locator. If you choose an XPath or DOM locator, you can verify that specific text appears at a specific location on the page relative to other UI components on the page.

CommandTargetValue
verifyText//table/tr/td/div/pThis is my text and it occurs right after the div inside the table.

Locating Elements

For many Selenium commands, a target is required. This target identifies an element in the content of the web application, and consists of the location strategy followed by the location in the format locatorType=location. The locator type can be omitted in many cases. The various locator types are explained below with examples for each.

Locating by Identifier

This is probably the most common method of locating elements and is the catch-all default when no recognized locator type is used. With this strategy, the first element with the id attribute value matching the location will be used. If no element has a matching id attribute, then the first element with a name attribute matching the location will be used.

For instance, your page source could have id and name attributes as follows:

  <html>
   <body>
    <form id="loginForm">
     <input name="username" type="text" />
     <input name="password" type="password" />
     <input name="continue" type="submit" value="Login" />
    </form>
   </body>
  <html>

The following locator strategies would return the elements from the HTML snippet above indicated by line number:

  • identifier=loginForm (3)
  • identifier=password (5)
  • identifier=continue (6)
  • continue (6)

Since the identifier type of locator is the default, the identifier= in the first three examples above is not necessary.

Locating by Id

This type of locator is more limited than the identifier locator type, but also more explicit. Use this when you know an element’s id attribute.

   <html>
    <body>
     <form id="loginForm">
      <input name="username" type="text" />
      <input name="password" type="password" />
      <input name="continue" type="submit" value="Login" />
      <input name="continue" type="button" value="Clear" />
     </form>
    </body>
   <html>
  • id=loginForm (3)

Locating by Name

The name locator type will locate the first element with a matching name attribute. If multiple elements have the same value for a name attribute, then you can use filters to further refine your location strategy. The default filter type is value (matching the value attribute).

   <html>
    <body>
     <form id="loginForm">
      <input name="username" type="text" />
      <input name="password" type="password" />
      <input name="continue" type="submit" value="Login" />
      <input name="continue" type="button" value="Clear" />
     </form>
   </body>
   <html>
  • name=username (4)
  • name=continue value=Clear (7)
  • name=continue Clear (7)
  • name=continue type=button (7)

Note: Unlike some types of XPath and DOM locators, the three types of locators above allow Selenium to test a UI element independent of its location on the page. So if the page structure and organization is altered, the test will still pass. You may or may not want to also test whether the page structure changes. In the case where web designers frequently alter the page, but its functionality must be regression tested, testing via id and name attributes, or really via any HTML property, becomes very important.

Locating by XPath

XPath is the language used for locating nodes in an XML document. As HTML can be an implementation of XML (XHTML), Selenium users can leverage this powerful language to target elements in their web applications. XPath extends beyond (as well as supporting) the simple methods of locating by id or name attributes, and opens up all sorts of new possibilities such as locating the third checkbox on the page.

One of the main reasons for using XPath is when you don’t have a suitable id or name attribute for the element you wish to locate. You can use XPath to either locate the element in absolute terms (not advised), or relative to an element that does have an id or name attribute. XPath locators can also be used to specify elements via attributes other than id and name.

Absolute XPaths contain the location of all elements from the root (html) and as a result are likely to fail with only the slightest adjustment to the application. By finding a nearby element with an id or name attribute (ideally a parent element) you can locate your target element based on the relationship. This is much less likely to change and can make your tests more robust.

Since only xpath locators start with “//”, it is not necessary to include the xpath= label when specifying an XPath locator.

   <html>
    <body>
     <form id="loginForm">
      <input name="username" type="text" />
      <input name="password" type="password" />
      <input name="continue" type="submit" value="Login" />
      <input name="continue" type="button" value="Clear" />
     </form>
   </body>
   <html>
  • xpath=/html/body/form[1] (3) - Absolute path (would break if the HTML was changed only slightly)
  • //form[1] (3) - First form element in the HTML
  • xpath=//form[@id='loginForm'] (3) - The form element with attribute named ‘id’ and the value ‘loginForm’
  • xpath=//form[input/@name='username'] (3) - First form element with an input child element with attribute named ‘name’ and the value ‘username’
  • //input[@name='username'] (4) - First input element with attribute named ‘name’ and the value ‘username’
  • //form[@id='loginForm']/input[1] (4) - First input child element of the form element with attribute named ‘id’ and the value ‘loginForm’
  • //input[@name='continue'][@type='button'] (7) - Input with attribute named ‘name’ and the value ‘continue’ and attribute named ‘type’ and the value ‘button’
  • //form[@id='loginForm']/input[4] (7) - Fourth input child element of the form element with attribute named ‘id’ and value ‘loginForm’

These examples cover some basics, but in order to learn more, the following references are recommended:

There are also a couple of very useful Firefox Add-ons that can assist in discovering the XPath of an element:

This is a simple method of locating a hyperlink in your web page by using the text of the link. If two links with the same text are present, then the first match will be used.

  <html>
   <body>
    <p>Are you sure you want to do this?</p>
    <a href="continue.html">Continue</a> 
    <a href="cancel.html">Cancel</a>
  </body>
  <html>
  • link=Continue (4)
  • link=Cancel (5)

Locating by DOM

The Document Object Model represents an HTML document and can be accessed using JavaScript. This location strategy takes JavaScript that evaluates to an element on the page, which can be simply the element’s location using the hierarchical dotted notation.

Since only dom locators start with “document”, it is not necessary to include the dom= label when specifying a DOM locator.

   <html>
    <body>
     <form id="loginForm">
      <input name="username" type="text" />
      <input name="password" type="password" />
      <input name="continue" type="submit" value="Login" />
      <input name="continue" type="button" value="Clear" />
     </form>
   </body>
   <html>
  • dom=document.getElementById('loginForm') (3)
  • dom=document.forms['loginForm'] (3)
  • dom=document.forms[0] (3)
  • document.forms[0].username (4)
  • document.forms[0].elements['username'] (4)
  • document.forms[0].elements[0] (4)
  • document.forms[0].elements[3] (7)

You can use Selenium itself as well as other sites and extensions to explore the DOM of your web application. A good reference exists on W3Schools.

Locating by CSS

CSS (Cascading Style Sheets) is a language for describing the rendering of HTML and XML documents. CSS uses Selectors for binding style properties to elements in the document. These Selectors can be used by Selenium as another locating strategy.

   <html>
    <body>
     <form id="loginForm">
      <input class="required" name="username" type="text" />
      <input class="required passfield" name="password" type="password" />
      <input name="continue" type="submit" value="Login" />
      <input name="continue" type="button" value="Clear" />
     </form>
   </body>
   <html>
  • css=form#loginForm (3)
  • css=input[name="username"] (4)
  • css=input.required[type="text"] (4)
  • css=input.passfield (5)
  • css=#loginForm input[type="button"] (7)
  • css=#loginForm input:nth-child(2) (5)

For more information about CSS Selectors, the best place to go is the W3C publication. You’ll find additional references there.

Implicit Locators

You can choose to omit the locator type in the following situations:

  • Locators without an explicitly defined locator strategy will default to using the identifier locator strategy. See Locating by Identifier_.

  • Locators starting with “//” will use the XPath locator strategy. See Locating by XPath_.

  • Locators starting with “document” will use the DOM locator strategy. See Locating by DOM_

Matching Text Patterns

Like locators, patterns are a type of parameter frequently required by Selenese commands. Examples of commands which require patterns are verifyTextPresent, verifyTitle, verifyAlert, assertConfirmation, verifyText, and verifyPrompt. And as has been mentioned above, link locators can utilize a pattern. Patterns allow you to describe, via the use of special characters, what text is expected rather than having to specify that text exactly.

There are three types of patterns: globbing, regular expressions, and exact.

Globbing Patterns

Most people are familiar with globbing as it is utilized in filename expansion at a DOS or Unix/Linux command line such as ls *.c. In this case, globbing is used to display all the files ending with a .c extension that exist in the current directory. Globbing is fairly limited.
Only two special characters are supported in the Selenium implementation:

* which translates to “match anything,” i.e., nothing, a single character, or many characters.

[ ] (character class) which translates to “match any single character found inside the square brackets.” A dash (hyphen) can be used as a shorthand to specify a range of characters (which are contiguous in the ASCII character set). A few examples will make the functionality of a character class clear:

[aeiou] matches any lowercase vowel

[0-9] matches any digit

[a-zA-Z0-9] matches any alphanumeric character

In most other contexts, globbing includes a third special character, the ?. However, Selenium globbing patterns only support the asterisk and character class.

To specify a globbing pattern parameter for a Selenese command, you can prefix the pattern with a glob: label. However, because globbing patterns are the default, you can also omit the label and specify just the pattern itself.

Below is an example of two commands that use globbing patterns. The actual link text on the page being tested was “Film/Television Department”; by using a pattern rather than the exact text, the click command will work even if the link text is changed to “Film & Television Department” or “Film and Television Department”. The glob pattern’s asterisk will match “anything or nothing” between the word “Film” and the word “Television”.

CommandTargetValue
clicklink=glob:Film*Television Department
verifyTitleglob:*Film*Television*

The actual title of the page reached by clicking on the link was “De Anza Film And Television Department - Menu”. By using a pattern rather than the exact text, the verifyTitle will pass as long as the two words “Film” and “Television” appear (in that order) anywhere in the page’s title. For example, if the page’s owner should shorten the title to just “Film & Television Department,” the test would still pass. Using a pattern for both a link and a simple test that the link worked (such as the verifyTitle above does) can greatly reduce the maintenance for such test cases.

Regular Expression Patterns

Regular expression patterns are the most powerful of the three types of patterns that Selenese supports. Regular expressions are also supported by most high-level programming languages, many text editors, and a host of tools, including the Linux/Unix command-line utilities grep, sed, and awk. In Selenese, regular expression patterns allow a user to perform many tasks that would be very difficult otherwise. For example, suppose your test needed to ensure that a particular table cell contained nothing but a number. regexp: [0-9]+ is a simple pattern that will match a decimal number of any length.

Whereas Selenese globbing patterns support only the * and [ ] (character class) features, Selenese regular expression patterns offer the same wide array of special characters that exist in JavaScript. Below are a subset of those special characters:

PATTERNMATCH
.any single character
[ ]character class: any single character that appears inside the brackets
*quantifier: 0 or more of the preceding character (or group)
+quantifier: 1 or more of the preceding character (or group)
?quantifier: 0 or 1 of the preceding character (or group)
{1,5}quantifier: 1 through 5 of the preceding character (or group)
|alternation: the character/group on the left or the character/group on the right
( )grouping: often used with alternation and/or quantifier

Regular expression patterns in Selenese need to be prefixed with either regexp: or regexpi:. The former is case-sensitive; the latter is case-insensitive.

A few examples will help clarify how regular expression patterns can be used with Selenese commands. The first one uses what is probably the most commonly used regular expression pattern–.* (“dot star”). This two-character sequence can be translated as “0 or more occurrences of any character” or more simply, “anything or nothing.” It is the equivalent of the one-character globbing pattern * (a single asterisk).

CommandTargetValue
clicklink=glob:Film*Television Department
verifyTitleregexp:.*Film.*Television.*

The example above is functionally equivalent to the earlier example that used globbing patterns for this same test. The only differences are the prefix (regexp: instead of glob:) and the “anything or nothing” pattern (.* instead of just *).

The more complex example below tests that the Yahoo! Weather page for Anchorage, Alaska contains info on the sunrise time:

CommandTargetValue
openhttp://weather.yahoo.com/forecast/USAK0012.html
verifyTextPresentregexp:Sunrise: *[0-9]{1,2}:[0-9]{2} [ap]m

Let’s examine the regular expression above one part at a time:

Sunrise: *The string Sunrise: followed by 0 or more spaces
[0-9]{1,2}1 or 2 digits (for the hour of the day)
:The character : (no special characters involved)
[0-9]{2}2 digits (for the minutes) followed by a space
[ap]m“a” or “p” followed by “m” (am or pm)

Exact Patterns

The exact type of Selenium pattern is of marginal usefulness. It uses no special characters at all. So, if you needed to look for an actual asterisk character (which is special for both globbing and regular expression patterns), the exact pattern would be one way to do that. For example, if you wanted to select an item labeled “Real *” from a dropdown, the following code might work or it might not. The asterisk in the glob:Real * pattern will match anything or nothing. So, if there was an earlier select option labeled “Real Numbers,” it would be the option selected rather than the “Real *” option.

CommandTargetValue
select//selectglob:Real *

In order to ensure that the “Real *” item would be selected, the exact: prefix could be used to create an exact pattern as shown below:

CommandTargetValue
select//selectexact:Real *

But the same effect could be achieved via escaping the asterisk in a regular expression pattern:

CommandTargetValue
select//selectregexp:Real \*

It’s rather unlikely that most testers will ever need to look for an asterisk or a set of square brackets with characters inside them (the character class for globbing patterns). Thus, globbing patterns and regular expression patterns are sufficient for the vast majority of us.

The “AndWait” Commands

The difference between a command and its AndWait alternative is that the regular command (e.g. click) will do the action and continue with the following command as fast as it can, while the AndWait alternative (e.g. clickAndWait) tells Selenium to wait for the page to load after the action has been done.

The AndWait alternative is always used when the action causes the browser to navigate to another page or reload the present one.

Be aware, if you use an AndWait command for an action that does not trigger a navigation/refresh, your test will fail. This happens because Selenium will reach the AndWait’s timeout without seeing any navigation or refresh being made, causing Selenium to raise a timeout exception.

The waitFor Commands in AJAX applications

In AJAX driven web applications, data is retrieved from server without refreshing the page. Using andWait commands will not work as the page is not actually refreshed. Pausing the test execution for a certain period of time is also not a good approach as web element might appear later or earlier than the stipulated period depending on the system’s responsiveness, load or other uncontrolled factors of the moment, leading to test failures. The best approach would be to wait for the needed element in a dynamic period and then continue the execution as soon as the element is found.

This is done using waitFor commands, as waitForElementPresent or waitForVisible, which wait dynamically, checking for the desired condition every second and continuing to the next command in the script as soon as the condition is met.

Sequence of Evaluation and Flow Control

When a script runs, it simply runs in sequence, one command after another.

Selenese, by itself, does not support condition statements (if-else, etc.) or iteration (for, while, etc.). Many useful tests can be conducted without flow control. However, for a functional test of dynamic content, possibly involving multiple pages, programming logic is often needed.

When flow control is needed, there are three options:

a) Run the script using Selenium-RC and a client library such as Java or PHP to utilize the programming language’s flow control features. b) Run a small JavaScript snippet from within the script using the storeEval command. c) Install the goto_sel_ide.js extension.

Most testers will export the test script into a programming language file that uses the Selenium-RC API (see the Selenium-IDE chapter). However, some organizations prefer to run their scripts from Selenium-IDE whenever possible (for instance, when they have many junior-level people running tests for them, or when programming skills are lacking). If this is your case, consider a JavaScript snippet or the goto_sel_ide.js extension.

Store Commands and Selenium Variables

You can use Selenium variables to store constants at the beginning of a script. Also, when combined with a data-driven test design (discussed in a later section), Selenium variables can be used to store values passed to your test program from the command-line, from another program, or from a file.

The plain store command is the most basic of the many store commands and can be used to simply store a constant value in a Selenium variable. It takes two parameters, the text value to be stored and a Selenium variable. Use the standard variable naming conventions of only alphanumeric characters when choosing a name for your variable.

CommandTargetValue
storepaul@mysite.org

Later in your script, you’ll want to use the stored value of your variable. To access the value of a variable, enclose the variable in curly brackets ({}) and precede it with a dollar sign like this.

CommandTargetValue
verifyText//div/p\${userName}

A common use of variables is for storing input for an input field.

CommandTargetValue
typeid=login\${userName}

Selenium variables can be used in either the first or second parameter and are interpreted by Selenium prior to any other operations performed by the command. A Selenium variable may also be used within a locator expression.

An equivalent store command exists for each verify and assert command. Here are a couple more commonly used store commands.

storeElementPresent

This corresponds to verifyElementPresent. It simply stores a boolean value–“true” or “false”–depending on whether the UI element is found.

storeText

StoreText corresponds to verifyText. It uses a locator to identify specific page text. The text, if found, is stored in the variable. StoreText can be used to extract text from the page being tested.

storeEval

This command takes a script as its first parameter. Embedding JavaScript within Selenese is covered in the next section. StoreEval allows the test to store the result of running the script in a variable.

JavaScript and Selenese Parameters

JavaScript can be used with two types of Selenese parameters: script and non-script (usually expressions). In most cases, you’ll want to access and/or manipulate a test case variable inside the JavaScript snippet used as a Selenese parameter. All variables created in your test case are stored in a JavaScript associative array. An associative array has string indexes rather than sequential numeric indexes. The associative array containing your test case’s variables is named storedVars. Whenever you wish to access or manipulate a variable within a JavaScript snippet, you must refer to it as storedVars[‘yourVariableName’].

JavaScript Usage with Script Parameters

Several Selenese commands specify a script parameter including assertEval, verifyEval, storeEval, and waitForEval. These parameters require no special syntax. A Selenium-IDE user would simply place a snippet of JavaScript code into the appropriate field, normally the Target field (because a script parameter is normally the first or only parameter).

The example below illustrates how a JavaScript snippet can be used to perform a simple numerical calculation:

CommandTargetValue
store10hits
storeXpathCount//blockquoteblockquotes
storeEvalstoredVars[‘hits’].storedVars[‘blockquotes’]paragraphs

This next example illustrates how a JavaScript snippet can include calls to methods, in this case the JavaScript String object’s toUpperCase method and toLowerCase method.

CommandTargetValue
storeEdith Whartonname
storeEvalstoredVars[‘name’].toUpperCase()uc
storeEvalstoredVars[‘name’].toUpperCase()lc

JavaScript Usage with Non-Script Parameters

JavaScript can also be used to help generate values for parameters, even when the parameter is not specified to be of type script.
However, in this case, special syntax is required–the entire parameter value must be prefixed by javascript{ with a trailing }, which encloses the JavaScript snippet, as in javascript{*yourCodeHere*}. Below is an example in which the type command’s second parameter value is generated via JavaScript code using this special syntax:

CommandTargetValue
storeleague of nationssearchString
typeqjavascript{storedVars[‘searchString’].toUpperCase()}

echo - The Selenese Print Command

Selenese has a simple command that allows you to print text to your test’s output. This is useful for providing informational progress notes in your test which display on the console as your test is running. These notes also can be used to provide context within your test result reports, which can be useful for finding where a defect exists on a page in the event your test finds a problem. Finally, echo statements can be used to print the contents of Selenium variables.

CommandTargetValue
echoTesting page footer now.
echoUsername is \${userName}

Alerts, Popups, and Multiple Windows

Suppose that you are testing a page that looks like this.

  <!DOCTYPE HTML>
  <html>
  <head>
    <script type="text/javascript">
      function output(resultText){
        document.getElementById('output').childNodes[0].nodeValue=resultText;
      }

      function show_confirm(){
        var confirmation=confirm("Chose an option.");
        if (confirmation==true){
          output("Confirmed.");
        }
        else{
          output("Rejected!");
        }
      }
      
      function show_alert(){
        alert("I'm blocking!");
        output("Alert is gone.");
      }
      function show_prompt(){
        var response = prompt("What's the best web QA tool?","Selenium");
        output(response);
      }
      function open_window(windowName){
        window.open("newWindow.html",windowName);
      }
      </script>
  </head>
  <body>

    <input type="button" id="btnConfirm" onclick="show_confirm()" value="Show confirm box" />
    <input type="button" id="btnAlert" onclick="show_alert()" value="Show alert" />
    <input type="button" id="btnPrompt" onclick="show_prompt()" value="Show prompt" />
    <a href="newWindow.html" id="lnkNewWindow" target="_blank">New Window Link</a>
    <input type="button" id="btnNewNamelessWindow" onclick="open_window()" value="Open Nameless Window" />
    <input type="button" id="btnNewNamedWindow" onclick="open_window('Mike')" value="Open Named Window" />

    <br />
    <span id="output">
    </span>
  </body>
  </html>

The user must respond to alert/confirm boxes, as well as moving focus to newly opened popup windows. Fortunately, Selenium can cover JavaScript pop-ups.

But before we begin covering alerts/confirms/prompts in individual detail, it is helpful to understand the commonality between them. Alerts, confirmation boxes and prompts all have variations of the following

CommandDescription
assertFoo(pattern)throws error if pattern doesn’t match the text of the pop-up
assertFooPresentthrows error if pop-up is not available
assertFooNotPresentthrows error if any pop-up is present
storeFoo(variable)stores the text of the pop-up in a variable
storeFooPresent(variable)stores the text of the pop-up in a variable and returns true or false

When running under Selenium, JavaScript pop-ups will not appear. This is because the function calls are actually being overridden at runtime by Selenium’s own JavaScript. However, just because you cannot see the pop-up doesn’t mean you don’t have to deal with it. To handle a pop-up, you must call its assertFoo(pattern) function. If you fail to assert the presence of a pop-up your next command will be blocked and you will get an error similar to the following [error] Error: There was an unexpected Confirmation! [Chose an option.]

Alerts

Let’s start with alerts because they are the simplest pop-up to handle. To begin, open the HTML sample above in a browser and click on the “Show alert” button. You’ll notice that after you close the alert the text “Alert is gone.” is displayed on the page. Now run through the same steps with Selenium IDE recording, and verify the text is added after you close the alert. Your test will look something like this:

CommandTargetValue
open/
clickbtnAlert
assertAlertI’m blocking!
verifyTextPresentAlert is gone.

You may be thinking “That’s odd, I never tried to assert that alert.” But this is Selenium-IDE handling and closing the alert for you. If you remove that step and replay the test you will get the following error [error] Error: There was an unexpected Alert! [I'm blocking!]. You must include an assertion of the alert to acknowledge its presence.

If you just want to assert that an alert is present but either don’t know or don’t care what text it contains, you can use assertAlertPresent. This will return true or false, with false halting the test.

Confirmations

Confirmations behave in much the same way as alerts, with assertConfirmation and assertConfirmationPresent offering the same characteristics as their alert counterparts. However, by default Selenium will select OK when a confirmation pops up. Try recording clicking on the “Show confirm box” button in the sample page, but click on the “Cancel” button in the popup, then assert the output text. Your test may look something like this:

CommandTargetValue
open/
clickbtnConfirm
chooseCancelOnNextConfirmation
assertConfirmationChoose an option.
verifyTextPresentRejected

The chooseCancelOnNextConfirmation function tells Selenium that all following confirmation should return false. It can be reset by calling chooseOkOnNextConfirmation.

You may notice that you cannot replay this test, because Selenium complains that there is an unhandled confirmation. This is because the order of events Selenium-IDE records causes the click and chooseCancelOnNextConfirmation to be put in the wrong order (it makes sense if you think about it, Selenium can’t know that you’re cancelling before you open a confirmation) Simply switch these two commands and your test will run fine.

Prompts

Prompts behave in much the same way as alerts, with assertPrompt and assertPromptPresent offering the same characteristics as their alert counterparts. By default, Selenium will wait for you to input data when the prompt pops up. Try recording clicking on the “Show prompt” button in the sample page and enter “Selenium” into the prompt. Your test may look something like this:

CommandTargetValue
open/
answerOnNextPromptSelenium!
clickid=btnPrompt
assertPromptWhat’s the best web QA tool?
verifyTextPresentSelenium!

If you choose cancel on the prompt, you may notice that answerOnNextPrompt will simply show a target of blank. Selenium treats cancel and a blank entry on the prompt basically as the same thing.

Debugging

Debugging means finding and fixing errors in your test case. This is a normal part of test case development.

We won’t teach debugging here as most new users to Selenium will already have some basic experience with debugging. If this is new to you, we recommend you ask one of the developers in your organization.

Breakpoints and Startpoints

The Sel-IDE supports the setting of breakpoints and the ability to start and stop the running of a test case, from any point within the test case. That is, one can run up to a specific command in the middle of the test case and inspect how the test case behaves at that point. To do this, set a breakpoint on the command just before the one to be examined.

To set a breakpoint, select a command, right-click, and from the context menu select Toggle Breakpoint. Then click the Run button to run your test case from the beginning up to the breakpoint.

It is also sometimes useful to run a test case from somewhere in the middle to the end of the test case or up to a breakpoint that follows the starting point.
For example, suppose your test case first logs into the website and then performs a series of tests and you are trying to debug one of those tests.
However, you only need to login once, but you need to keep rerunning your tests as you are developing them. You can login once, then run your test case from a startpoint placed after the login portion of your test case. That will prevent you from having to manually logout each time you rerun your test case.

To set a startpoint, select a command, right-click, and from the context menu select Set/Clear Start Point. Then click the Run button to execute the test case beginning at that startpoint.

Stepping Through a Testcase

To execute a test case one command at a time (“step through” it), follow these steps:

  1. Start the test case running with the Run button from the toolbar.

  2. Immediately pause the executing test case with the Pause button.

  3. Repeatedly select the Step button.

Find Button

The Find button is used to see which UI element on the currently displayed webpage (in the browser) is used in the currently selected Selenium command.
This is useful when building a locator for a command’s first parameter (see the section on :ref:locators <locators-section> in the Selenium Commands chapter). It can be used with any command that identifies a UI element on a webpage, i.e. click, clickAndWait, type, and certain assert and verify commands, among others.

From Table view, select any command that has a locator parameter. Click the Find button.
Now look on the webpage: There should be a bright green rectangle enclosing the element specified by the locator parameter.

Page Source for Debugging

Often, when debugging a test case, you simply must look at the page source (the HTML for the webpage you’re trying to test) to determine a problem. Firefox makes this easy. Simply right-click the webpage and select ‘View->Page Source.
The HTML opens in a separate window. Use its Search feature (Edit=>Find) to search for a keyword to find the HTML for the UI element you’re trying to test.

Alternatively, select just that portion of the webpage for which you want to see the source. Then right-click the webpage and select View Selection Source. In this case, the separate HTML window will contain just a small amount of source, with highlighting on the portion representing your selection.

Locator Assistance

Whenever Selenium-IDE records a locator-type argument, it stores additional information which allows the user to view other possible locator-type arguments that could be used instead. This feature can be very useful for learning more about locators, and is often needed to help one build a different type of locator than the type that was recorded.

This locator assistance is presented on the Selenium-IDE window as a drop-down list accessible at the right end of the Target field (only when the Target field contains a recorded locator-type argument).
Below is a snapshot showing the contents of this drop-down for one command. Note that the first column of the drop-down provides alternative locators, whereas the second column indicates the type of each alternative.

Selenium Locator Assistance

Writing a Test Suite

A test suite is a collection of test cases which is displayed in the leftmost pane in the IDE.
The test suite pane can be manually opened or closed via selecting a small dot halfway down the right edge of the pane (which is the left edge of the entire Selenium-IDE window if the pane is closed).

The test suite pane will be automatically opened when an existing test suite is opened or when the user selects the New Test Case item from the File menu. In the latter case, the new test case will appear immediately below the previous test case.

Selenium-IDE also supports loading pre-existing test cases by using the File -> Add Test Case menu option. This allows you to add existing test cases to a new test suite.

A test suite file is an HTML file containing a one-column table. Each cell of each row in thesection contains a link to a test case. The example below is of a test suite containing four test cases:

<html>
    <head>
        <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
        <title>Sample Selenium Test Suite</title>
    </head>
    <body>
        <table cellpadding="1" cellspacing="1" border="1">
            <thead>
                <tr><td>Test Cases for De Anza A-Z Directory Links</td></tr>
            </thead>
        <tbody>
            <tr><td><a href="./a.html">A Links</a></td></tr>
            <tr><td><a href="./b.html">B Links</a></td></tr>
            <tr><td><a href="./c.html">C Links</a></td></tr>
            <tr><td><a href="./d.html">D Links</a></td></tr>
        </tbody>
        </table>
    </body>
</html>

Note: Test case files should not have to be co-located with the test suite file that invokes them. And on Mac OS and Linux systems, that is indeed the case. However, at the time of this writing, a bug prevents Windows users from being able to place the test cases elsewhere than with the test suite that invokes them.

User Extensions

User extensions are JavaScript files that allow one to create his or her own customizations and features to add additional functionality. Often this is in the form of customized commands although this extensibility is not limited to additional commands.

There are a number of useful extensions_ created by users.

IMPORTANT: THIS SECTION IS OUT OF DATE–WE WILL BE REVISING THIS SOON.

.. _goto_sel_ide.js extension:

Perhaps the most popular of all Selenium-IDE extensions is one which provides flow control in the form of while loops and primitive conditionals. This extension is the goto_sel_ide.js_. For an example of how to use the functionality provided by this extension, look at the page_ created by its author.

To install this extension, put the pathname to its location on your computer in the Selenium Core extensions field of Selenium-IDE’s Options=>Options=>General tab.

Selenium IDE Extensions Install

After selecting the OK button, you must close and reopen Selenium-IDE in order for the extensions file to be read. Any change you make to an extension will also require you to close and reopen Selenium-IDE.

Information on writing your own extensions can be found near the bottom of the Selenium Reference_ document.

Sometimes it can prove very useful to debug step by step Selenium IDE and your User Extension. The only debugger that appears able to debug XUL/Chrome based extensions is Venkman which is supported in Firefox until version 32 included. The step by step debug has been verified to work with Firefox 32 and Selenium IDE 2.9.0.

Format

Format, under the Options menu, allows you to select a language for saving and displaying the test case. The default is HTML.

If you will be using Selenium-RC to run your test cases, this feature is used to translate your test case into a programming language. Select the language, e.g. Java, PHP, you will be using with Selenium-RC for developing your test programs. Then simply save the test case using File=>Export Test Case As. Your test case will be translated into a series of functions in the language you choose. Essentially, program code supporting your test is generated for you by Selenium-IDE.

Also, note that if the generated code does not suit your needs, you can alter it by editing a configuration file which defines the generation process.
Each supported language has configuration settings which are editable. This is under the Options=>Options=>Formats tab.

Executing Selenium-IDE Tests on Different Browsers

While Selenium-IDE can only run tests against Firefox, tests developed with Selenium-IDE can be run against other browsers, using a simple command-line interface that invokes the Selenium-RC server. This topic is covered in the :ref:Run Selenese tests <html-suite> section on Selenium-RC chapter. The -htmlSuite command-line option is the particular feature of interest.

Troubleshooting

Below is a list of image/explanation pairs which describe frequent sources of problems with Selenium-IDE:

Table view is not available with this format.

This message can be occasionally displayed in the Table tab when Selenium IDE is launched. The workaround is to close and reopen Selenium IDE. See issue 1008. for more information. If you are able to reproduce this reliably then please provide details so that we can work on a fix.


error loading test case: no command found

You’ve used File=>Open to try to open a test suite file. Use File=>Open Test Suite instead.

An enhancement request has been raised to improve this error message. See issue 1010.


Selenium IDE Trouble Timing

This type of error may indicate a timing problem, i.e., the element specified by a locator in your command wasn’t fully loaded when the command was executed. Try putting a pause 5000 before the command to determine whether the problem is indeed related to timing. If so, investigate using an appropriate waitFor* or *AndWait command before the failing command.


Selenium IDE Trouble Param

Whenever your attempt to use variable substitution fails as is the case for the open command above, it indicates that you haven’t actually created the variable whose value you’re trying to access. This is sometimes due to putting the variable in the Value field when it should be in the Target field or vice versa. In the example above, the two parameters for the store command have been erroneously placed in the reverse order of what is required. For any Selenese command, the first required parameter must go in the Target field, and the second required parameter (if one exists) must go in the Value field.


error loading test case: [Exception… “Component returned failure code: 0x80520012 (NS_ERROR_FILE_NOT_FOUND) [nsIFileInputStream.init]” nresult: “0x80520012 (NS_ERROR_FILE_NOT_FOUND)” location: “JS frame :: chrome://selenium-ide/content/file-utils.js :: anonymous :: line 48” data: no]

One of the test cases in your test suite cannot be found. Make sure that the test case is indeed located where the test suite indicates it is located. Also, make sure that your actual test case files have the .html extension both in their filenames, and in the test suite file where they are referenced.

An enhancement request has been raised to improve this error message. See issue 1011.


Selenium IDE Trouble Extension

Your extension file’s contents have not been read by Selenium-IDE. Be sure you have specified the proper pathname to the extensions file via Options=>Options=>General in the Selenium Core extensions field. Also, Selenium-IDE must be restarted after any change to either an extensions file or to the contents of the Selenium Core extensions field.

5 - Le serveur Remote WebDriver

Le serveur fonctionnera toujours sur la machine avec le navigateur que vous souhaitez tester. Le serveur peut être utilisé à partir de la ligne de commande ou via du code configuration.

Démarrage du serveur à partir de la ligne de commande

Une fois que vous avez téléchargé selenium-server-standalone-{VERSION}.jar, placez-le sur l’ordinateur avec le navigateur que vous souhaitez tester. Ensuite, à partir de le répertoire avec le pot, exécutez ce qui suit:

java -jar selenium-server-standalone-{VERSION}.jar

Considérations pour exécuter le serveur

L’appelant doit terminer correctement chaque session, appelant soit Selenium#stop() ou WebDriver#quit.

Le ‘Selenium-Server’ conserve des journaux en mémoire pour chaque session en cours, qui sont effacés lorsque Selenium#stop() ou WebDriver#quit est appelé. Si vous oubliez de mettre fin à ces sessions, votre serveur peut perdre de la mémoire. Si vous gardez des sessions extrêmement longues, vous devrez probablement arrêtez/quittez de temps en temps (ou augmentez la mémoire avec l’option -Xmx jvm).

Timeouts (à partir de la version 2.21)

Le serveur a deux délais d’expiration différents, qui peuvent être définis comme suit:

java -jar selenium-server-standalone-{VERSION}.jar -timeout=20 -browserTimeout=60
  • browserTimeout
    • Contrôle la durée pendant laquelle le navigateur est autorisé à se bloquer (valeur en secondes).
  • timeout
    • Contrôle la durée pendant laquelle le client est autorisé à partir   avant la récupération de la session (valeur en secondes).

La propriété système selenium.server.session.timeout n’est plus pris en charge à partir de la version 2.21.

Veuillez noter que le browserTimeout est conçu comme un mécanisme de temporisation de sauvegarde lorsque le mécanisme de temporisation ordinaire échoue, qui devrait être utilisé principalement dans des environnements de grille/serveur pour garantir que les processus bloqués/perdus ne restent pas trop longtemps, polluant l’environnement d’exécution.

Configuration du serveur par programmation

En théorie, le processus est aussi simple que de mapper le DriverServlet à une URL, mais il est également possible d’héberger la page dans un format léger conteneur, tel que Jetty configuré entièrement en code. Étapes à suivre pour ce faire suivre.

  • Téléchargez selenium-server.zip et décompressez.
  • Placez les JAR sur le CLASSPATH.
  • Créez une nouvelle classe appelée AppServer. Ici, j’utilise Jetty, vous devrez donc télécharger cela aussi:
import org.mortbay.jetty.Connector;
import org.mortbay.jetty.Server;
import org.mortbay.jetty.nio.SelectChannelConnector;
import org.mortbay.jetty.security.SslSocketConnector;
import org.mortbay.jetty.webapp.WebAppContext;

import javax.servlet.Servlet;
import java.io.File;

import org.openqa.selenium.remote.server.DriverServlet;

public class AppServer {
  private Server server = new Server();

  public AppServer() throws Exception {
    WebAppContext context = new WebAppContext();
    context.setContextPath("");
    context.setWar(new File("."));
    server.addHandler(context);

    context.addServlet(DriverServlet.class, "/wd/*");

    SelectChannelConnector connector = new SelectChannelConnector();
    connector.setPort(3001);
    server.addConnector(connector);

    server.start();
  }
}

6 - Grid 3

Most of the documentation found in this section is still in English. Please note we are not accepting pull requests to translate this content as translating documentation of legacy components does not add value to the community nor the project.

Selenium Grid est un serveur proxy intelligent qui permet aux tests Selenium d’acheminer des commandes vers des instances de navigateur Web distantes. Son objectif est de fournir un moyen simple d’exécuter des tests en parallèle sur plusieurs machines.

Avec Selenium Grid, un serveur fait office de concentrateur qui achemine les commandes de test au format JSON à un ou plusieurs nœuds de grille enregistrés. Les tests contactent le concentrateur pour obtenir l’accès aux instances de navigateur distantes. Le concentrateur dispose d’une liste de serveurs enregistrés auxquels il donne accès, et permet le contrôle de ces instances.

Selenium Grid nous permet d’effectuer des tests en parallèle sur plusieurs machines, et pour gérer les différentes versions et configurations de navigateur de manière centralisée (au lieu de dans chaque test individuel).

La grille de sélénium n’est pas une solution miracle. Il résout un sous-ensemble de problèmes de délégation et de distribution communs, mais ne gèrera par exemple pas votre infrastructure, et pourrait ne pas répondre à vos besoins spécifiques.

6.1 - Composant

Grid 3 Components

Hub

  • Intermédiaire et gestionnaire
  • Accepte les demandes d’exécution de tests
  • Prend les instructions du client et les exécute à distance sur les nœuds
  • Gère les threads

Un Hub est un point central où tous vos tests sont envoyés. Chaque grille de sélénium se compose exactement d’un concentrateur. Le hub doit être accessible des clients respectifs (c.-à-d. serveur CI, machine développeur, etc.) Le concentrateur connectera un ou plusieurs nœuds auquel les tests seront délégués.

Nodes

  • Où vivent les navigateurs
  • S’enregistre auprès du hub et communique ses capacités
  • Reçoit les demandes du hub et les exécute

Nodes sont différentes instances de Selenium qui exécutera des tests sur des systèmes informatiques individuels. Il peut y avoir plusieurs nœuds dans une grille. Les machines qui sont des nœuds n’ont pas besoin d’être la même plate-forme ou avoir la même sélection de navigateur que celle du concentrateur ou des autres nœuds. Un nœud sous Windows peut avoir la capacité de offrant Internet Explorer comme option de navigateur, alors que cela ne serait pas possible sur Linux ou Mac.

6.2 - Mettre en place votre propre Grid

Pour utiliser la grille de Selenium, vous devez maintenir votre propre infrastructure pour les nœuds. Comme cela peut être un effort lourd et intense en temps, de nombreuses organisations utilisent des fournisseurs IaaS comme Amazon EC2 et Google Compute pour fournir cette infrastructure.

D’autres options incluent l’utilisation de fournisseurs tels que Sauce Labs ou Testing Bot qui fournissent un Selenium Grid en tant que service dans le cloud. Il est certainement possible d’exécuter également des nœuds sur votre propre matériel. Ce chapitre abordera en détail l’option de faire fonctionner votre propre grille, complet avec sa propre infrastructure de nœuds.

Quick start

Cet exemple vous montrera comment démarrer le Selenium 2 Grid Hub, et enregistrer à la fois un nœud WebDriver et un nœud hérité Selenium 1 RC. Nous vous montrerons également comment appeler la grille depuis Java. Le hub et les nœuds sont montrés ici fonctionnant sur la même machine, mais bien sûr, vous pouvez copier le serveur autonome de Selenium sur plusieurs machines.

Le package selenium-server-standalone comprend le concentrateur, WebDriver et l’héritage RC nécessaires pour exécuter la grille, ant n’est plus requis. Vous pouvez télécharger le selenium-server-standalone.jar depuis https://selenium.dev/downloads/.

Étape 1: démarrez le hub

Le Hub est le point central qui recevra les demandes de test et les distribuer aux bons nœuds. La distribution se fait sur la base des capacités, ce qui signifie un test nécessitant un ensemble de capacités ne sera distribué qu’aux nœuds offrant cet ensemble ou sous-ensemble de capacités.

Parce que les capacités souhaitées d’un test sont exactement ce que son nom l’indique, desired, le concentrateur ne peut pas garantir qu’il localisera un nœud correspondant parfaitement à l’ensemble de capacités souhaité.

Ouvrez une invite de commande et accédez au répertoire dans lequel vous avez copié le fichier selenium-server-standalone.jar. Vous démarrez le hub en passant le drapeau -role hub sur le serveur autonome:

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

The Hub will listen to port 4444 by default. You can view the status of the hub by opening a browser window and navigating to http://localhost:4444/grid/console.

Le concentrateur écoutera le port 4444 par défaut. Vous pouvez afficher l’état du concentrateur en ouvrant une fenêtre de navigateur et en accédant à http://localhost:4444/grid/console.

Pour modifier le port par défaut, vous pouvez ajouter le drapeau facultatif -port avec un entier représentant le port à écouter lorsque vous exécutez la commande. En outre, toutes les autres options que vous voyez dans le fichier de configuration JSON (voir ci-dessous) sont des drapeaux de ligne de commande possibles.

Vous pouvez certainement vous en tirer avec la simple commande ci-dessus, mais si vous avez besoin d’une configuration plus avancée, vous pouvez également spécifier un fichier de configuration au format JSON, pour plus de commodité, pour configurer le concentrateur au démarrage. Vous pouvez le faire comme ceci:

java -jar selenium-server-standalone.jar -role hub -hubConfig hubConfig.json -debug

Ci-dessous, vous verrez un exemple de fichier hubConfig.json. Nous verrons plus en détail comment fournir les fichiers de configuration de noeud à l’étape 2.

{
  "_comment" : "Configuration for Hub - hubConfig.json",
  "host": ip,
  "maxSession": 5,
  "port": 4444,
  "cleanupCycle": 5000,
  "timeout": 300000,
  "newSessionWaitTimeout": -1,
  "servlets": [],
  "prioritizer": null,
  "capabilityMatcher": "org.openqa.grid.internal.utils.DefaultCapabilityMatcher",
  "throwOnCapabilityNotPresent": true,
  "nodePolling": 180000,
  "platform": "WINDOWS"}

Étape 2: démarrez les nœuds

Que vous souhaitiez exécuter une grille avec de nouvelles fonctionnalités WebDriver, ou une grille avec la fonctionnalité Selenium 1 RC, Ou les deux à la fois, vous utilisez le même fichier selenium-server-standalone.jar pour démarrer les nœuds:

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

Si un port n’est pas spécifié via l’indicateur -port, un port libre sera choisi. Vous pouvez exécuter plusieurs nœuds sur une seule machine mais si vous le faites, vous devez être conscient des ressources mémoire de votre système et des problèmes avec les captures d’écran si vos tests les prennent.

Configuration du nœud avec options

Comme mentionné, pour une compatibilité ascendante Les rôles “wd” et “rc” sont toujours un sous-ensemble valide du rôle “node”. Mais ces rôles limitent les types de connexions distantes à leur API correspondante, tandis que “node” permet à la fois les connexions à distance RC et WebDriver.

Passer les propriétés JVM (en utilisant l’indicateur -D avant l’argument -jar) sur la ligne de commande, et ceux-ci seront ramassés et propagés aux nœuds:

-Dwebdriver.chrome.driver=chromedriver.exe

Configuration de Node avec JSON

Vous pouvez également démarrer des nœuds de grille configurés avec un fichier de configuration JSON

java -Dwebdriver.chrome.driver=chromedriver.exe -jar selenium-server-standalone.jar -role node -nodeConfig node1Config.json

Et voici un exemple de fichier nodeConfig.json:

{
  "capabilities": [
    {
      "browserName": "firefox",
      "acceptSslCerts": true,
      "javascriptEnabled": true,
      "takesScreenshot": false,
      "firefox_profile": "",
      "browser-version": "27",
      "platform": "WINDOWS",
      "maxInstances": 5,
      "firefox_binary": "",
      "cleanSession": true
    },
    {
      "browserName": "chrome",
      "maxInstances": 5,
      "platform": "WINDOWS",
      "webdriver.chrome.driver": "C:/Program Files (x86)/Google/Chrome/Application/chrome.exe"
    },
    {
      "browserName": "internet explorer",
      "maxInstances": 1,
      "platform": "WINDOWS",
      "webdriver.ie.driver": "C:/Program Files (x86)/Internet Explorer/iexplore.exe"
    }
  ],
  "configuration": {
    "_comment" : "Configuration for Node",
    "cleanUpCycle": 2000,
    "timeout": 30000,
    "proxy": "org.openqa.grid.selenium.proxy.WebDriverRemoteProxy",
    "port": 5555,
    "host": ip,
    "register": true,
    "hubPort": 4444,
    "maxSession": 5
  }
}

Une note sur le drapeau -host

Pour le concentrateur et le nœud, si l’indicateur -host n’est pas spécifié, 0.0.0.0 sera utilisé par défaut. Cela se liera à tous les interfaces IPv4 publiques (sans boucle) de la machine. Si vous avez un spécial configuration réseau ou tout composant créant des interfaces réseau supplémentaires, il est conseillé de définir le drapeau -host avec une valeur qui concentrateur/node accessible depuis une autre machine.

Spécification du port

Le port TCP/IP par défaut utilisé par le concentrateur est 4444. Si vous devez changer le port veuillez utiliser les configurations mentionnées ci-dessus.

Dépannage

Utilisation du fichier journal

Pour un dépannage avancé, vous pouvez spécifier un fichier journal pour enregistrer les messages système. Démarrez le concentrateur ou le nœud Selenium GRID avec l’argument -log. Veuillez voir l’exemple ci-dessous:

java -jar selenium-server-standalone.jar -role hub -log log.txt

Utilisez votre éditeur de texte préféré pour ouvrir le fichier journal (log.txt dans l’exemple ci-dessus) pour trouver Journaux “ERREUR” si vous rencontrez des problèmes.

Utilisation de l’argument -debug

Vous pouvez également utiliser l’argument -debug pour imprimer les journaux de débogage sur la console. Démarrez Selenium Grid Hub ou Node avec l’argument -debug. S’il te plait regarde l’exemple ci-dessous:

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

Attention

La grille de Selenium doit être protégée des accès externes à l’aide de autorisations de pare-feu.

Le fait de ne pas protéger votre grille peut entraîner un ou plusieurs des événements suivants:

  • Vous fournissez un accès ouvert à votre infrastructure Grid
  • Vous autorisez des tiers à accéder aux applications et fichiers Web internes
  • Vous autorisez des tiers à exécuter des binaires personnalisés

Voir cet article de blog sur Detectify, qui donne une bonne aperçu de l’utilisation abusive d’une grille exposée publiquement: Ne laissez pas votre grille grande ouverte.

Docker Selenium

Docker fournit un moyen pratique de fournir et dimensionner l’infrastructure Selenium Grid dans une unité appelée conteneur. Les conteneurs sont des unités logicielles standardisées qui contiennent tout le nécessaire pour exécuter l’application souhaitée, y compris toutes les dépendances, dans un environnement fiable et reproductible sur différentes machines.

Le projet Selenium gère un ensemble d’images Docker que vous pouvez télécharger et exécutez pour obtenir une grille de travail rapidement opérationnelle. Les nœuds sont disponibles pour Firefox et Chrome. Vous trouverez tous les détails sur la configuration d’un réseau dans le Docker Selenium dépôt.

Prérequis

La seule exigence pour exécuter une grille est d’avoir Docker installé et fonctionnel. Installer Docker.