This section contains description of FluentLenium features which may be useful during writing tests.

Table of Contents

Window actions

Window actions are available on window() method.

Few additional methods are available from ones inherited from Selenium, like clickAndOpenNew, openNewAndSwitch, clickAndCloseCurrent and switchToLast

  • clickAndOpenNew

    You should use this method when click action invoked on element should end up with new window opened. This is just about waiting for new window to open, this method is not opening new browser window.

  • openNewAndSwitch

    This method opens new window and switches the context to newly opened one.

  • clickAndCloseCurrent

    You should use this method when click action invoked on element should end up with current window closed. This is just about waiting for new window to close, this method is not closing browser window.

  • switchToLast

    This method switch to the last window. If argument is provided with the name of a window, ensure that the last window has this name.

You can also maximize browser window:


Keyboard and Mouse actions

Advanced keyboard and mouse actions are available using keyboard() and mouse() in FluentTest class or element.

Drag and drop

Drag and drop can be done using MouseElementActions

public void dragAndDrop(){
        FluentWebElement source=el("#source");
        FluentWebElement target=el("target");
        MouseElementActions actions=new MouseElementActions(getDriver(),source.getElement());


Selenium has a driver wrapper named EventFiringWebDriver that is able to generate events and register listeners.

FluentLenium brings Events Annotations and a Listener API to register those listeners easily.


You can use annotations from package to register any method as an event listener.

Annotations can be used in a test class.

public void afterClickOn(FluentWebElement element){
        System.out.println("Element Clicked: "+element);

Annotations related to a WebElement can also be used in a component class.

public class SomeComponent {
    private WebElement element;

    public SomeComponent(WebElement element) {
        this.element = element;

    public SomeComponent click() {;
        return this;

    private void afterClickOn() {
        System.out.println("Element Clicked: " + this);

public class SomeTest extends FluentTest {

    private SomeComponent clickComponent;
    private SomeComponent otherComponent;

    public void test() {;
        // @AfterClickOn annotated method will be invoked for clickComponent instance only.

Annotations may be also be configured with a priority, so that annotations with higher priority value will be executed first.

public class SomeComponent {
    private WebElement element;

    public SomeComponent(WebElement element) {
        this.element = element;

    public SomeComponent click() {;
        return this;

    private void afterClickOnHigher() {
        System.out.println("Element Clicked, with higher priority: " + this);

    private void afterClickOnLower() {
        System.out.println("Element Clicked, with lower priority: " + this);

The default priority value of all event annotations are 0.

Using event annotations events will be registered with a default annotation based listener implementation which cannot be overridden.

Listener API

You can also register events through API using events() method.

events().afterClickOn(new ElementListener(){
public void on(FluentWebElement element,WebDriver driver){
        System.out.println("Element Clicked: "+element);

        el("button").click(); // This will call the listener.

This integrates nicely with Java 8 lambdas:

events().afterClickOn((element,driver)->System.out.println("Element Clicked: "+element));

        el("button").click(); // This will call the listener.

The events() method provides a more flexible way of registering listeners as it allows one to register their custom listener implementations, and the examples above demonstrate it well.

To also be able to handle the priority of these custom listeners one must also implement the ListenerPriority interface, so that listeners get sorted by their priority.

private final class AfterClick implements ListenerPriority, ElementListener {

    public void on(FluentWebElement element, WebDriver driver) {
        System.out.println("A really awesome click happened!");

    public int getPriority() {
        return 1;

Combining the two solutions

Annotation and Listener API based listeners can be used along with each other. If you define priorities for the event annotations, and you don’t implement ListenerPriority in case of custom listeners, the custom listeners will be executed last among the registered listeners.


It’s possible to add some behavior for any element without changing its code by using hooks.

FluentLenium is shipped with the Wait hook.

// This call will wait for ajax-element to be clickable.

// Hooks can be chained and their effect will be associated.

// Options can be given to hooks.

Hook Annotations

Hooks can also be mapped to annotation, like @Wait for the WaitHook class.

You can place a hook annotation on injected Field, Fluent Test class or Page.

The hook will be effective for a field if annotation is present on it. All fields from a Fluent Test class if it’s present on a Fluent Test class and all fields from a Page if it’s present on a Page.

Page build through Injection recursively inherits hooks from parent pages and fluent test.

If you need to disable all inherited hooks in particular Page or Field, you should use @NoHook annotation, or call noHook() function on the element, with no parameter to remove all hooks, of with parameters to remove only given hooks.

It’s also possible to use the generic @Hook annotation to enable a hook class.

public class MyTest extends FluentTest {
    FluentWebElement waitElement;

    FluentWebElement waitAndAnotherHookElement;

    FluentWebElement noHookElement;

    MyPage page;

public class MyPage extends FluentPage {


Custom hook

It’s possible to implement your own hook by extending BaseHook or BaseFluentHook.

Let’s implement Example hook by writing a configurable message before and after click.

  • Create the hook option class. It should be a JavaBean containing configuration options for the hook.
public class ExampleHookOptions {
    private String message = "ExampleHook";

    public String getMessage() {
        return message;

    public void setMessage(String message) {
        this.message = message;
  • Create the hook by extending BaseHook or BaseFluentHook. It should at least have a public constructor matching parent class, and generic type should be a hook option class.
public class ExampleHook extends BaseHook<ExampleHookOptions> {
    public ExampleHook(FluentControl control, ComponentInstantiator instantiator, Supplier<WebElement> elementSupplier, Supplier<ElementLocator> locatorSupplier, ExampleHookOptions options) {
        super(control, instantiator, elementSupplier, locatorSupplier, options);

  • Create the hook annotation. Annotation parameters should match options found in hook option class.

@Target({ElementType.FIELD, ElementType.TYPE})
public @interface Example {
    String message() default "";
  • Add a constructor to hook options class that will support loading of hook annotation into hook option class.
public class ExampleHookOptions {
    public ExampleHookOptions(Example annotation) {
        message = annotation.message();

    public ExampleHookOptions() {
        // Default constructor
  • Override methods from WebElement or ElementLocator in the hook implementation class to add desired behavior.
public void click(){
        System.out.println(getOptions().getMessage()+": before click!");;
        System.out.println(getOptions().getMessage()+": after click!");

Example sources are available on github .

You may also read sources for @Wait hook to read how it’s implemented.

Javascript execution

If you need to execute some javascript, just call executeScript with your script as parameter. For example, if you have a javascript method called change and you want to call it just add this in your test:


You can either execute javascript with arguments, with async executeAsyncScript, and retrieve the result.


CSS injection

It is possible to manipulate the CSS styling of pages via calling css().

It can be called from a FluentWebElement, FluentDriver and any of the test adapter implementations.

You can either inject an explicitly defined CSS styling:

css().inject("#location {\ndisplay: none\n}");

or you can inject the content of a CSS resource from the project classpath:


Taking ScreenShots and HTML Dumps

You can take a ScreenShot and an HTML Dump of the browser.


The file will be named using the current timestamp. You can of course specify a path and a name using:


Screenshot and HTML Dump can be automatically performed on test fail using configuration properties.

When using AUTOMATIC_ON_FAIL with JUnit, you should use custom @After annotation from io.fluentlenium.adapter.junit package for screenshot and HTML dump to be performed just after an exception occurs, before methods annotated with @After invocation.


If you want to switch the Selenium webDriver to an iframe (see this Selenium FAQ) , you can just call the method switchTo() :

To switch to the default context:

        switchToDefault(); // Alternative method

To switch to the iframe selected:



If you want manage alert (see this Selenium FAQ) ,

When an alert box pops up, click on “OK”:


When an alert box pops up, click on “Cancel”:


Entering an input value in prompt:


Chromium API

FluentLenium gives you an opportunity to run Chrome DevTools commands in your tests using simple Chromium API.

What is Chromium API?

Chromium API gives you an easy access to theChrome DevTools Protocol. This was possible by extending commands supported by Selenium with two new endpoints to directly call the DevTools API:

  • /session/:sessionId/chromium/send_command_and_get_result
  • /session/:sessionId/chromium/send_comman

It will allow you to easily automate Chrome browser beyond the standard WebDriver protocol.

Why you may want to use it?

With the DevTool protocol will be able to achieve more powerful interactions with the browser compare to WebDriver. Taking full page screenshot or clearing cookies for every domain are just two examples how you could use it.

How to use it?

  public class DuckDuckGoChromiumApiTest extends FluentTest {

    private Response response;

    public WebDriver newWebDriver() {
        return new ChromeDriver();

    public void resultPageUrlShouldContainSearchQueryName() {
        String searchPhrase = "searchPhrase";
        String duckDuckUrl = "";

        getChromiumApi().sendCommand("Page.navigate", ImmutableMap.of("url", duckDuckUrl));
        getChromiumApi().sendCommand("Input.insertText", ImmutableMap.of("text", searchPhrase));
        getChromiumApi().sendCommand("Input.dispatchKeyEvent", sendEnterKeyEventParams());
        response = getChromiumApi().sendCommandAndGetResponse("Page.getNavigationHistory", ImmutableMap.of());


    private Map<String, String> sendEnterKeyEventParams() {
        return ImmutableMap.of("type", "char", "text", "\r");

    private void assertIsPhrasePresentInTheResultsPageUrl(String searchPhrase) {

Performance Timing API

FluentLenium provides an API for retrieving the performance timing metrics based on the PerformanceTiming interface defined by W3C .

The main interface for this API is PerformanceTiming from which you can query individual metric values and metrics in bulk as well. Most methods return long values querying the window.performance.timing.<metric> Javascript attribute, except the ones that are explicitly stated in the W3C documentation that they may have other type of values as well.

They can be retrieved via FluentTest and its other framework specific variants:

public class SomeTest extends FluentTest {

    private Homepage homepage;

    public void test() {
        //Get metric via parameterized method
        long loadEventEnd = performanceTiming().getEventValue(PerformanceTimingEvent.LOAD_EVENT_END);

        //The same as the previous call but the value is converted to the given time unit
        long loadEventEndInSeconds = performanceTiming().getEventValue(PerformanceTimingEvent.LOAD_EVENT_END, TimeUnit.SECONDS);

        //This is a convenience method for calling performanceTiming().getEventValue(PerformanceTimingEvent.DOM_COMPLETE);
        long domComplete = performanceTiming().domComplete();

        //The same as the previous call but the value is converted to the given time unit
        long domCompleteInSeconds = performanceTiming().domComplete(TimeUnit.SECONDS);


    and via `FluentPage` as well:

public class Homepage extends FluentPage {

    public long getDomComplete() {
        return performanceTiming().domComplete();

    public long getLoadEventEnd() {
        return performanceTiming().getEventValue(PerformanceTimingEvent.LOAD_EVENT_END);

Each method returning a specific metric execute a separate Javascript command.

There is another way to get metrics, specifically to get all metrics in a single object called PerformanceTimingMetrics. This returns the object returned by the window.performance.timing Javascript attribute.

public void test(){
        //This returns the metrics by default in default milliseconds
        PerformanceTimingMetrics metrics=performanceTiming().getMetrics();
        long domComplete=metrics.getDomComplete();

        //This returns a new metrics object that will return the values in the set time unit
        long domCompleteInSeconds=metricsInSeconds.getDomComplete();

In this case only a single Javascript command is executed for performanceTiming().getMetrics(), getDomComplete() ( actually any method) on this object returns the saved value, and none of the getter methods execute any additional Javascript command.

It is important to note the the implementations of both the PerformanceTiming and PerformanceTimingMetrics interfaces provided by FluentLenium return handle the navigationStart attribute as zero and every other metric is calculated and returned relative to navigationStart.

Before retrieving a performance timing metrics value make sure that the page where you query it loaded completely. In case when navigation happens to a specific URL, or bz some action performed on the page make sure in your test that the page where it navigates to loads completely. Otherwise certain metrics might not have been registered until that moment.

You can find some examples in the FluentLenium project for how you can use these features.