(For more resources related to this topic, see here.)
This article is completely devoted to testing in Groovy. Testing is probably the most important activity that allows us to produce better software and make our users happier. The Java space has countless tools and frameworks that can be used for testing our software. In this article, we will direct our focus on some of those frameworks and how they can be integrated with Groovy. We will discuss not only unit testing techniques, but also integration and load testing strategies.
Starting from the king of all testing frameworks, JUnit and its seamless Groovy integration, we move to explore how to test:
SOAP and REST web services
Code that interacts with databases
The web application interface using Selenium
The article also covers Behavior Driven Development (BDD) with Spock, advanced web service testing using soapUI, and load testing using JMeter.
Unit testing Java code with Groovy
One of the ways developers start looking into the Groovy language and actually using it is by writing unit tests. Testing Java code with Groovy makes the tests less verbose and it's easier for the developers that clearly express the intent of each test method.
Thanks to the Java/Groovy interoperability, it is possible to use any available testing or mocking framework from Groovy, but it's just simpler to use the integrated JUnit based test framework that comes with Groovy.
In this recipe, we are going to look at how to test Java code with Groovy.
Getting ready
This recipe requires a new Groovy project that we will use again in other recipes of this article. The project is built using Gradle and contains all the test cases required by each recipe.
Let's create a new folder called groovy-test and add a build.gradle file to it. The build file will be very simple:
apply plugin: 'groovy' apply plugin: 'java' repositories {
mavenCentral() maven {
url 'https://oss.sonatype.org' + '/content/repositories/snapshots'
}
}
dependencies {
compile 'org.codehaus.groovy:groovy-all:2.1.6' testCompile
'junit:junit:4.+'
}
The standard source folder structure has to be created for the project. You can use the following commands to achieve this:
mkdir -p src/main/groovy
mkdir -p src/test/groovy
mkdir -p src/main/java
Verify that the project builds without errors by typing the following command in your shell:
gradle clean build
How to do it...
To show you how to test Java code from Groovy, we need to have some Java code first!
So, this recipe's first step is to create a simple Java class called StringUtil, which we will test in Groovy. The class is fairly trivial, and it exposes only one method that concatenates String passed in as a List:
package org.groovy.cookbook;
import java.util.List;
public class StringUtil
{
public String concat(List<String> strings, String separator)
{
StringBuilder sb = new StringBuilder(); String sep = "";
for(String s: strings) { sb.append(sep).append(s);
sep = separator;
}
return sb.toString();
}
}
Note that the class has a specific package, so don't forget to create the appropriate folder structure for the package when placing the class in the src/main/java folder.
Run the build again to be sure that the code compiles.
Now, add a new test case in the src/test/groovy folder:
package org.groovy.cookbook.javatesting
import org.groovy.cookbook.StringUtil
class JavaTest extends GroovyTestCase
{
def stringUtil = new StringUtil() void testConcatenation()
{
def result = stringUtil.concat(['Luke', 'John'], '-')
assertToString('Luke-John', result)
}
void testConcatenationWithEmptyList()
{
def result = stringUtil.concat([], ',') assertEquals('', result)
}
}
Again, pay attention to the package when creating the test case class and create the package folder structure.
Run the test by executing the following Gradle command from your shell:
gradle clean build
Gradle should complete successfully with the BUILD SUCCESSFUL message.
Now, add a new test method and run the build again:
void testConcatenationWithNullShouldReturnNull()
{
def result = stringUtil.concat(null, ',') assertEquals('', result)
}
This time the build should fail:
3 tests completed, 1 failed :test FAILED
FAILURE: Build failed with an exception.
Fix the test by adding a null check to the Java code as the first statement of the concat method:
if (strings == null)
{
return "";
}
Run the build again to verify that it is now successful.
How it works...
The test case shown at step 3 requires some comments. The class extends GroovyTestCase is a base test case class that facilitates the writing of unit tests by adding several helper methods to the classes extending it. When a test case extends from GroovyTestCase, each test method name must start with test and the return type must be void. It is possible to use the JUnit 4.x @Test annotation, but, in that case, you don't have to extend from GroovyTestCase.
The standard JUnit assertions (such as assertEquals and assertNull) are directly available in the test case (without explicit import) plus some additional assertion methods are added by the super class. The test case at step 3 uses assertToString to verify that a String matches the expected result. There are other assertions added by GroovyTestCase, such as assertArrayEquals, to check that two arrays contain the same values, or assertContains to assert that an array contains a given element.
There's more...
The GroovyTestCase class also offers an elegant method to test for expected exceptions. Let's add the following rule to the concat method:
if (separator.length() != 1) {
throw new IllegalArgumentException(
"The separator must be one char long");
}
Place the separator length check just after the null check for the List. Add the following new test method to the test case:
void testVerifyExceptionOnWrongSeparator()
{
shouldFail IllegalArgumentException,
{
stringUtil(['a', 'b'], ',,')
}
shouldFail IllegalArgumentException, { stringUtil(['c', 'd'], '')
}
}
The shouldFail method takes a closure that is executed in the context of a try-catch block. We can also specify the expected exception in the shouldFail method. The shouldFail method makes testing for exceptions very elegant and simple to read.
See also
http://junit.org/
http://groovy.codehaus.org/api/groovy/util/GroovyTestCase.html
Testing SOAP web services
This recipe shows you how to use the JUnit 4 annotations instead of the JUnit 3 API offered by GroovyTestCase.
Getting ready
For this recipe, we are going to use a publicly available web service, the US holiday date web service hosted at http://www.holidaywebservice.com. The WSDL of the service can be found on the /Holidays/US/Dates/USHolidayDates.asmx?WSDL path. We have already encountered this service in the Issuing a SOAP request and parsing a response recipe in Article 8, Working with Web Services in Groovy. Each service operation simply returns the date of a given US holiday such as Easter or Christmas.
How to do it...
We start from the Gradle build that we created in the Unit testing Java code with Groovy recipe.
Add the following dependency to the dependencies section of the build.gradle file:
testCompile 'com.github.groovy-wslite:groovy-wslite:0.8.0'
Let's create a new unit test for verifying one of the web service operations. As usual, the test case is created in the src/test/groovy/org/groovy/cookbook folder:
package org.groovy.cookbook.soap
import static org.junit.Assert.* import org.junit.Test
import wslite.soap.*
class SoapTest {
...
}
Add a new test method to the body of the class:
@Test
void testMLKDay() {
def baseUrl = 'http://www.holidaywebservice.com'
def service = '/Holidays/US/Dates/USHolidayDates.asmx?WSDL'
def client = new SOAPClient("${baseUrl}${service}")
def baseNS = 'http://www.27seconds.com/Holidays/US/Dates/'
def action = "${baseNS}GetMartinLutherKingDay"
def response = client.send(SOAPAction: action) { body
{
GetMartinLutherKingDay('>gradle -Dtest.single=SoapTest clean test
How it works...
The test code creates a new SOAPClient with the URI of the target web service. The request is created using Groovy's MarkupBuilder. The body closure (and if needed, also the header closure) is passed to the MarkupBuilder for the SOAP message creation. The assertion code gets the result from the response, which is automatically parsed by XMLSlurper, allowing easy access to elements of the response such as the header or the body elements. In the previous test, we simply check that the returned Martin Luther King day matches with the expected one for the year 2013.
There's more...
If you require more control over the content of the SOAP request, the SOAPClient also supports sending the SOAP envelope as a String, such as in this example:
def response = client.send (
"""<?xml version='1.0' encoding='UTF-8'?> <soapenv:Envelope>
<soapenv:Header/>
<soapenv:Body>
<dat:GetMartinLutherKingDay>
<dat:year>2013</dat:year>
</dat:GetMartinLutherKingDay>
</soapenv:Body>
</soapenv:Envelope>
"""
)
Replace the call to the send method in step 3 with the one above and run your test again.
See also
https://github.com/jwagenleitner/groovy-wslite
http://www.holidaywebservice.com
Testing RESTful services
This recipe is very similar to the previous recipe Testing SOAP web services, except that it shows how to test a RESTful service using Groovy and JUnit.
Getting ready
For this recipe, we are going to use a test framework aptly named Rest-Assured. This framework is a simple DSL for testing and validating REST services returning either JSON or XML.
Before we start to delve into the recipe, we need to start a simple REST service for testing purposes. We are going to use the Ratpack framework. The test REST service, which we will use, exposes three APIs to fetch, add, and delete books from a database using JSON as lingua franca.
For the sake of brevity, the code for the setup of this recipe is available in the rest-test folder of the companion code for this article. The code contains the Ratpack server, the domain objects, Gradle build, and the actual test case that we are going to analyze in the next section.
How to do it...
The test case takes care of starting the Ratpack server and execute the REST requests.
Here is the RestTest class located in src/test/groovy/org/groovy/ cookbok/rest folder:
src/test/groovy/org/groovy/ cookbok/rest folder:
package org.groovy.cookbook.rest
import static com.jayway.restassured.RestAssured.*
import static com.jayway.restassured.matcher.
RestAssuredMatchers.*
import static org.hamcrest.Matchers.* import static org.junit.Assert.*
import groovy.json.JsonBuilder
import org.groovy.cookbook.server.* import org.junit.AfterClass
import org.junit.BeforeClass import org.junit.Test
class RestTest {
static server
final static HOST = 'http://localhost:5050'
@BeforeClass
static void setUp() { server = App.init() server.startAndWait()
}
@AfterClass
static void tearDown() { if(server.isRunning()) {
server.stop()
}
}
@Test
void testGetBooks() { expect().
body('author',
hasItems('Ian Bogost', 'Nate Silver')). when().get("${HOST}/api/books")
}
@Test
void testGetBook() { expect().
body('author', is('Steven Levy')). when().get("${HOST}/api/books/5")
}
@Test
void testPostBook() { def book = new Book()
book.author = 'Haruki Murakami' book.date = '2012-05-14'
book.title = 'Kafka on the shore' JsonBuilder jb = new JsonBuilder()
jb.content = book
given(). content(jb.toString()).
expect().body('id', is(6)). when().post("${HOST}/api/books/new")
}
@Test
void testDeleteBook() { expect().statusCode(200).
when().delete("${HOST}/api/books/1") expect().body('id', not(hasValue(1))).
when().get("${HOST}/api/books")
}
}
Build the code and execute the test from the command line by typing:
gradle clean test
How it works...
The JUnit test has a @BeforeClass annotated method, executed at the beginning of the unit test, that starts the Ratpack server and the associated REST services. The @AfterClass annotated method, on the contrary, shuts down the server when the test is over.
The unit test has four test methods. The first one, testGetBooks executes a GET request against the server and retrieves all the books.
The rather readable DSL offered by the Rest-Assured framework should be easy to follow. The expect method starts building the response expectation returned from the get method. The actual assert of the test is implemented via a Hamcrest matcher (hence the static org.hamcrest.Matchers.* import in the test). The test is asserting that the body of the response contains two books that have the author named Ian Bogost or Greg Grandin. The get method hits the URL of the embedded Ratpack server, started at the beginning of the test.
The testGetBook method is rather similar to the previous one, except that it uses the is matcher to assert the presence of an author on the returned JSON message.
The testPostBook tests that the creation of a new book is successful. First, a new book object is created and transformed into a JSON object using JsonBuilder. Instead of the expect method, we use the given method to prepare the POST request. The given method returns a RequestSpecification to which we assign the newly created book and finally, invoke the post method to execute the operation on the server. As in our book database, the biggest identifier is 8, the new book should get the id 9, which we assert in the test.
The last test method (testDeleteBook) verifies that a book can be deleted. Again we use the expect method to prepare the response, but this time we verify that the returned HTTP status code is 200 (for deletion) upon deleting a book with the id 1. The same test also double-checks that fetching the full list of books does not contain the book with id equals to 1.
See also
https://code.google.com/p/rest-assured/
https://code.google.com/p/hamcrest/
https://github.com/ratpack/ratpack
Writing functional tests for web applications
If you are developing web applications, it is of utmost importance that you thoroughly test them before allowing user access. Web GUI testing can require long hours of very expensive human resources to repeatedly exercise the application against a varied list of input conditions. Selenium, a browser-based testing and automation framework, aims to solve these problems for software developers, and it has become the de facto standard for web interface integration and functional testing.
Selenium is not just a single tool but a suite of software components, each catering to different testing needs of an organization. It has four main parts:
Selenium Integrated Development Environment (IDE): It is a Firefox add-on that you can only use for creating relatively simple test cases and test suites.
Selenium Remote Control (RC): It also known as Selenium 1, is the first Selenium tool that allowed users to use programming languages in creating complex tests.
WebDriver: It is the newest Selenium implementation that allows your test scripts to communicate directly to the browser, thereby controlling it from the OS level.
Selenium Grid: It is a tool that is used with Selenium RC to execute parallel tests across different browsers and operating systems.
Since 2008, Selenium RC and WebDriver are merged into a single framework to form Selenium 2. Selenium 1 refers to Selenium RC.
This recipe will show you how to write a Selenium 2 based test using HtmlUnitDriver. HtmlUnitDriver is the fastest and most lightweight implementation of WebDriver at the moment. As the name suggests, it is based on HtmlUnit, a relatively old framework for testing web applications.
The main disadvantage of using this driver instead of a WebDriver implementation that "drives" a real browser is the JavaScript support. None of the popular browsers use the JavaScript engine used by HtmlUnit (Rhino). If you test JavaScript using HtmlUnit, the results may divert considerably from those browsers. Still, WebDriver and HtmlUnit can be used for fast paced testing against a web interface, leaving more JavaScript intensive tests to other, long running, WebDriver implementations that use specific browsers.
Getting ready
Due to the relative complexity of the setup required to demonstrate the steps of this recipe, it is recommended that the reader uses the code that comes bundled with this recipe. The code is located in the selenium-test located in the code directory for this article. The source code, as other recipes in this article, is built using Gradle and has a standard structure, containing application code and test code.
The web application under test is very simple. It is composed of two pages. A welcome page that looks similar to the following screenshot:
And a single field test form page:
The Ratpack framework is utilized to run the fictional web application and serve the HTML pages along with some JavaScript and CSS.
How to do it...
The following steps will describe the salient points of Selenium testing with Groovy.
Let's open the build.gradle file. We are interested in the dependencies required to execute the tests:
testCompile group: 'org.seleniumhq.selenium', name:
'selenium-htmlunit-driver',
version: '2.32.0'
testCompile group: 'org.seleniumhq.selenium', name: 'selenium-support',
version: '2.9.0'
Let's open the test case, SeleniumTest.groovy located in test/groovy/org/ groovy/cookbook/selenium:
test/groovy/org/ groovy/cookbook/selenium:
package org.groovy.cookbook.selenium import static org.junit.Assert.*
import org.groovy.cookbook.server.* import org.junit.AfterClass
import org.junit.BeforeClass import org.junit.Test
import org.openqa.selenium.By import org.openqa.selenium.WebDriver
import org.openqa.selenium.WebElement
import org.openqa.selenium.htmlunit.HtmlUnitDriver import
org.openqa.selenium.support.ui.WebDriverWait
import com.google.common.base.Function class SeleniumTest {
static server
static final HOST = 'http://localhost:5050' static HtmlUnitDriver driver
@BeforeClass
static void setUp() { server = App.init() server.startAndWait()
driver = new HtmlUnitDriver(true)
}
@AfterClass
static void tearDown() {
if (server.isRunning()) { server.stop()
}
}
@Test
void testWelcomePage() { driver.get(HOST) assertEquals('welcome',
driver.title)
}
@Test
void testFormPost() {
driver.get("${HOST}/form") assertEquals('test form', driver.title)
WebElement input = driver.findElement(By.name('username'))
input.sendKeys('test')
input.submit()
WebDriverWait wait = new WebDriverWait(driver, 4) wait.until
ExpectedConditions.
presenceOfElementLocated(By.className('hello'))
assertEquals('oh hello,test', driver.title)
}
}
How it works...
The test case initializes the Ratpack server and the HtmlUnit driver by passing true to the HtmlUnitDriver instance. The boolean parameter in the constructor indicates whether the driver should support JavaScript.
The first test, testWelcomePage, simply verifies that the title of the website's welcome page is as expected. The get method executes an HTTP GET request against the URL specified in the method, the Ratpack server in our test.
The second test, testFormPost, involves the DOM manipulation of a form, its submission, and waiting for an answer from the server.
The Selenium API should be fairly readable. For a start, the test checks that the page containing the form has the expected title. Then the element named username (a form field) is selected, populated, and finally submitted. This is how the HTML looks for the form field:
<input type="text" name="username" placeholder="Your username">
The test uses the findElement method to select the input field. The method expects a By object that is essentially a mechanism to locate elements within a document. Elements can be identified by name, id, text link, tag name, CSS selector, or XPath expression.
The form is submitted via AJAX. Here is part of the JavaScript activated by the form submission:
complete:function(xhr, status) {
if (status === 'error' || !xhr.responseText) { alert('error')
} else {
document.title = xhr.responseText
jQuery(e.target).
replaceWith('<p class="hello">' + xhr.responseText +
'</p>')
}
}
After the form submission, the DOM of the page is manipulated to change the page title of the form page and replace the form DOM element with a message wrapped in a paragraph element.
To verify that the DOM changes have been applied, the test uses the WebDriverWait class to wait until the DOM is actually modified and the element with the class hello appears on the page. The WebDriverWait is instantiated with a four seconds timeout.
This recipe only scratches the surface of the Selenium 2 framework's capabilities, but it should get you started to implement your own integration and functional test.
See also
http://docs.seleniumhq.org/
http://htmlunit.sourceforge.net/
Writing behavior-driven tests with Groovy
Behavior Driven Development, or simply BDD, is a methodology where QA, business analysts, and marketing people could get involved in defining the requirements of a process in a common language. It could be considered an extension of Test Driven Development, although is not a replacement. The initial motivation for BDD stems from the perplexity of business people (analysts, domain experts, and so on) to deal with "tests" as these seem to be too technical. The employment of the word "behaviors" in the conversation is a way to engage the whole team.
BDD states that software tests should be specified in terms of the desired behavior of the unit. The behavior is expressed in a semi-formal format, borrowed from user story specifications, a format popularized by agile methodologies.
For a deeper insight into the BDD rationale, it is highly recommended to read the original paper from Dan North available at http://dannorth.net/introducing-bdd/.
Spock is one of the most widely used frameworks in the Groovy and Java ecosystem that allows the creation of BDD tests in a very intuitive language and facilitates some common tasks such as mocking and extensibility. What makes it stand out from the crowd is its beautiful and highly expressive specification language. Thanks to its JUnit runner, Spock is compatible with most IDEs, build tools, and continuous integration servers.
In this recipe, we are going to look at how to implement both a unit test and a web integration test using Spock.
Getting ready
This recipe has a slightly different setup than most of the recipes in this book, as it resorts to an existing web application source code, freely available on the Internet. This is the Spring Petclinic application provided by SpringSource as a demonstration of the latest Spring framework features and configurations. The web application works as a pet hospital and most of the interactions are typical CRUD operations against certain entities (veterinarians, pets, pet owners). The Petclinic application is available in the groovy-test/spock/web folder of the companion code for this article.
All Petclinic's original tests have been converted to Spock. Additionally we created a simple integration test that uses Spock and Selenium to showcase the possibilities offered by the integration of the two frameworks.
As usual, the recipe uses Gradle to build the reference web application and the tests.
The Petclinic web application can be started by launching the following Gradle command in your shell from the groovy-test/spock/web folder:
gradle tomcatRunWar
If the application starts without errors, the shell should eventually display the following message:
The Server is running at http://localhost:8080/petclinic
Take some time to familiarize yourself with the Petclinic application by directing your browser to http://localhost:8080/petclinic and browsing around the website:
How to do it...
The following steps will describe the key concepts for writing your own behavior-driven unit and integration tests:
Let's start by taking a look at the dependencies required to implement a Spock-based test suite:
testCompile 'org.spockframework:spock-core:0.7-groovy-2.0'
testCompile group: 'org.seleniumhq.selenium',
name: 'selenium-java', version: '2.16.1'
testCompile group: 'junit', name: 'junit', version: '4.10' testCompile
'org.hamcrest:hamcrest-core:1.2'
testRuntime 'cglib:cglib-nodep:2.2' testRuntime 'org.objenesis:objenesis:1.2'
This is what a BDD unit test for the application's business logic looks like:
package org.springframework.samples.petclinic.model import spock.lang.*
class OwnerTest extends Specification { def "test pet and owner" () {
given:
def p = new Pet() def o = new Owner()
when:
p.setName("fido")
then:
o.getPet("fido") == null o.getPet("Fido") == null
when:
o.addPet(p)
then: o.getPet("fido").equals(p) o.getPet("Fido").equals(p)
}
}
The test is named OwnerTest.groovy, and it is available in the spock/web/src/ test folder of the main groovy-test project that comes with this article.
The third test in this recipe mixes Spock and Selenium, the web testing framework already discussed in Writing functional tests for web applications recipe:
package org.cookbook.groovy.spock
import static java.util.concurrent.TimeUnit.SECONDS import
org.openqa.selenium.By
import org.openqa.selenium.WebElement
import org.openqa.selenium.htmlunit.HtmlUnitDriver import
spock.lang.Shared
import spock.lang.Specification
class HomeSpecification extends Specification {
static final HOME = 'http://localhost:9966/petclinic'
@Shared
def driver = new HtmlUnitDriver(true)
def setup() { driver.manage().timeouts().implicitlyWait 10, SECONDS
}
def 'user enters home page'() { when:
driver.get(HOME)
then:
driver.title == 'PetClinic :: ' + 'a Spring Framework demonstration'
}
def 'user clicks on menus'() {
when:
driver.get(HOME)
def vets = driver.findElement(By.linkText('Veterinarians')) vets.click()
then: driver.currentUrl ==
'http://localhost:9966/petclinic/vets.html'
}
}
The test above is available in the spock/specs/src/test folder of the accompanying project.
How it works...
The first step of this recipe lays out the dependencies required to set up a Spock-based BDD test suite. Spock requires Java 5 or higher, and it's pretty picky with regard to the matching Groovy version to use. In the case of this recipe, as we are using Groovy 2.x, we set the dependency to the 0.7-groovy-2.0 version of the Spock framework.
The full build.gradle file is located in the spock/specs folder of the recipe's code.
The first test case demonstrated in the recipe is a direct conversion of a JUnit test written by Spring for the Petclinic application. This is the original test written in Java:
public class OwnerTests { @Test
public void testHasPet() { Owner owner = new Owner();Pet fido = new Pet(); fido.setName("Fido");
assertNull(owner.getPet("Fido"));
assertNull(owner.getPet("fido"));
owner.addPet(fido);
assertEquals(fido, owner.getPet("Fido")); assertEquals(fido,
owner.getPet("fido"));
}
}
All we need to import in the Spock test is spock.lang.* that contains the most important types for writing specifications. A Spock test extends from spock.lang.Specification. The name of a specification normally relates to the system or system operation under test. In the case of the Groovy test at step 2, we reused the original Java test name, but it would have been better renamed to something more meaningful for a specification such as OwnerPetSpec. The class Specification exposes a number of practical methods for implementing specifications. Additionally, it tells JUnit to run the specification with Sputnik, Spock's own JUnit runner. Thanks to Sputnik, Spock specifications can be executed by all IDEs and build tools.
Following the class definition, we have the feature method:
def 'test pet and owner'() {
...
}
Feature methods lie at the core of a specification. They contain the description of the features (properties, aspects) that define the system that is under specification test. Feature methods are conventionally named with String literals: it's a good idea to choose meaningful names for feature methods.
In the test above, we are testing that: given two entities, pet and owner, the getPet method of the owner instance will return null until the pet is not assigned to the owner, and that the getPet method will accept both "fido" and "Fido" in order to verify the ownership.
Conceptually, a feature method consists of four phases:
Set up the feature's fixture
Provide an input to the system under specification (stimulus)
Describe the response expected from the system Clean up
Whereas the first and last phases are optional, the stimulus and response phases are always present and may occur more than once.
Each phase is defined by blocks: blocks are defined by a label and extend to the beginning of the next block or the end of the method. In the test at step 2, we can see 3 types of blocks:
In the given block, data gets initialized
The when and then blocks always occur together. They describe a stimulus and the expected response. Whereas when blocks may contain arbitrary code; then blocks are restricted to conditions, exception checking, interactions, and variable definitions.
The first test case has two when/then pairs. A pet is assigned the name "fido", and the test verifies that calling getPet on an owner object only returns something if the pet is actually "owned" by the owner.
The second test is slightly more complex because it employs the Selenium framework to execute a web integration test with a BDD flavor. The test is located in the groovy-test/spock/specs/src/test folder. You can launch it by typing gradle test from the groovy-test/spock/specs folder. The test takes care of starting the web container and run the application under test, Petclinic. The test starts by defining a shared Selenium driver marked with the @Shared annotation, which is visible by all the feature methods. The first feature method simply opens the Petclinic main page and checks that the title matches the specification. The second feature method uses the Selenium API to select a link, click on it, and verify that the link brings the user to the right page. The verification is performed against the currentUrl of the browser that is expected to match the URL of the link we clicked on.
See also
http://dannorth.net/introducing-bdd/
http://en.wikipedia.org/wiki/Behavior-driven_development
https://code.google.com/p/spock/
https://github.com/spring-projects/spring-petclinic/
Read more
United States
Great Britain
India
Germany
France
Canada
Russia
Spain
Brazil
Australia
Singapore
Canary Islands
Hungary
Ukraine
Luxembourg
Estonia
Lithuania
South Korea
Turkey
Switzerland
Colombia
Taiwan
Chile
Norway
Ecuador
Indonesia
New Zealand
Cyprus
Denmark
Finland
Poland
Malta
Czechia
Austria
Sweden
Italy
Egypt
Belgium
Portugal
Slovenia
Ireland
Romania
Greece
Argentina
Netherlands
Bulgaria
Latvia
South Africa
Malaysia
Japan
Slovakia
Philippines
Mexico
Thailand
