selenium-cucumber-js

JavaScript browser automation framework using official selenium-webdriver and cucumber-js.

If you prefer to work with puppeteer check out puppeteer-cucumber-js

Table of Contents

• Installation
• Usage
  • Options
    • Configuration file
  • Feature files
  • Step definitions
  • Page objects
  • Shared objects
  • Helpers
  • Visual Comparison
  • Before/After hooks
  • Reports
  • How to debug
  • Directory structure
• Demo
• Bugs
• Contributing
• Troubleshooting
  • IntelliJ Cucumber Plugin

Installation

npm install selenium-cucumber-js --save-dev

Usage

node ./node_modules/selenium-cucumber-js/index.js -s ./step-definitions

Options

-h, --help                          output usage information
-V, --version                       output the version number
-s, --steps <path>                  path to step definitions. defaults to ./step-definitions
-p, --pageObjects <path>            path to page objects. defaults to ./page-objects
-o, --sharedObjects [paths]         path to shared objects (repeatable). defaults to ./shared-objects
-b, --browser <path>                name of browser to use. defaults to chrome
-k, --browser-teardown <optional>   browser teardown strategy after every scenario (always, clear, none). defaults to "always"
-r, --reports <path>                output path to save reports. defaults to ./reports
-d, --disableLaunchReport           disable the auto opening the browser with test report
-j, --junit <path>                  output path to save junit-report.xml defaults to ./reports
-t, --tags <tagName>                name of tag to run
-f, --featureFile <path>            a specific feature file to run
-x, --timeOut <n>                   steps definition timeout in milliseconds. defaults to 10 seconds
-n, --noScreenshot                  disable auto capturing of screenshots when an error is encountered

By default tests are run using Google Chrome, to run tests using another browser supply the name of that browser along with the -b switch. Available options are:

To use your own driver, create a customDriver.js file in the root of your project and provide the filename with the -b switch.

Configuration file

Configuration options can be set using a selenium-cucumber-js.json file at the root of your project. The JSON keys use the "long name" from the command line options. For example the following duplicates default configuration:

{
    "steps": "./step-definitions",
    "pageObjects": "./page-objects",
    "sharedObjects": "./shared-objects",
    "reports": "./reports",
    "browser": "chrome",
    "timeout": 10000
}

Whereas the following would set configuration to match the expected directory structure of IntelliJ's Cucumber plugin, and make default timeout one minute. Note that the default browser has not been overridden and will remain 'chrome'.

{
    "steps": "./features/step_definitions",
    "pageObjects": "./features/page_objects",
    "sharedObjects": "./features/shared_objects",
    "reports": "./features/reports",
    "timeout": 60000
}

Feature files

A feature file is a Business Readable, Domain Specific Language file that lets you describe software’s behavior without detailing how that behavior is implemented. Feature files are written using the Gherkin syntax and must live in a folder named features within the root of your project.

# ./features/google-search.feature

Feature: Searching for vote cards app
  As an internet user
  In order to find out more about the itunes vote cards app
  I want to be able to search for information about the itunes vote cards app

  Scenario: Google search for vote cards app
    When I search Google for "itunes vote cards app"
    Then I should see some results

Browser teardown strategy

The browser automatically closes after each scenario to ensure the next scenario uses a fresh browser environment. But you can change this behavior with the "-k" or the "--browser-teardown" parameter.

Step definitions

Step definitions act as the glue between features files and the actual system under test.

To avoid confusion always return a JavaScript promise your step definition in order to let cucumber know when your task has completed.

// ./step-definitions/google-search-steps.js

module.exports = function () {

    this.Then(/^I should see some results$/, function () {

        // driver wait returns a promise so return that
        return driver.wait(until.elementsLocated(by.css('div.g')), 10000).then(function(){

            // return the promise of an element to the following then.
            return driver.findElements(by.css('div.g'));
        })
        .then(function (elements) {

            // verify this element has children
            expect(elements.length).to.not.equal(0);
        });
    });
};

The following variables are available within the Given(), When() and Then() functions:

Page objects

Page objects are accessible via a global page object and are automatically loaded from ./page-objects (or the path specified using the -p switch). Page objects are exposed via a camel-cased version of their filename, for example ./page-objects/google-search.js becomes page.googleSearch. You can also use subdirectories, for example ./page-objects/dir/google-search.js becomes page.dir.googleSearch.

Page objects also have access to the same runtime variables available to step definitions.

An example page object:

// ./page-objects/google-search.js

module.exports = {

    url: 'http://www.google.co.uk',

    elements: {
        searchInput: by.name('q'),
        searchResultLink: by.css('div.g > h3 > a')
    },

    /**
     * enters a search term into Google's search box and presses enter
     * @param {string} searchQuery
     * @returns {Promise} a promise to enter the search values
     */
    performSearch: function (searchQuery) {

        var selector = page.googleSearch.elements.searchInput;

        // return a promise so the calling function knows the task has completed
        return driver.findElement(selector).sendKeys(searchQuery, selenium.Key.ENTER);
    }
};

And its usage within a step definition:

// ./step-definitions/google-search-steps.js
this.When(/^I search Google for "([^"]*)"$/, function (searchQuery) {

    return helpers.loadPage('http://www.google.com').then(function() {

        // use a method on the page object which also returns a promise
        return page.googleSearch.performSearch(searchQuery);
    })
});

Shared objects

Shared objects allow you to share anything from test data to helper methods throughout your project via a global shared object. Shared objects are automatically loaded from ./shared-objects (or the path specified using the -o switch) and made available via a camel-cased version of their filename, for example ./shared-objects/test-data.js becomes shared.testData. You can also use subdirectories, for example ./shared-objects/dir/test-data.js becomes shared.dir.testData.

Shared objects also have access to the same runtime variables available to step definitions.

An example shared object:

// ./shared-objects/test-data.js

module.exports = {
    username: "import-test-user",
    password: "import-test-pa**word"
}

And its usage within a step definition:

module.exports = function () {

    this.Given(/^I am logged in"$/, function () {

        driver.findElement(by.name('usn')).sendKeys(shared.testData.username);
        driver.findElement(by.name('pass')).sendKeys(shared.testData.password);
    });
};

Helpers

selenium-cucumber-js contains a few helper methods to make working with selenium a bit easier, those methods are:

// Load a URL, returning only when the <body> tag is present
helpers.loadPage('http://www.google.com');

// get the value of a HTML attribute
helpers.getAttributeValue('body', 'class');

// get a list of elements matching a query selector who's inner text matches param.
helpers.getElementsContainingText('nav[role="navigation"] ul li a', 'Safety Boots');

// get first elements matching a query selector who's inner text matches textToMatch param
helpers.getFirstElementContainingText('nav[role="navigation"] ul li a', 'Safety Boots');

// click element(s) that are not visible (useful in situations where a menu needs a hover before a child link appears)
helpers.clickHiddenElement('nav[role="navigation"] ul li a','Safety Boots');

// wait until a HTML attribute equals a particular value
helpers.waitUntilAttributeEquals('html', 'data-busy', 'false', 5000);

// wait until a HTML attribute exists
helpers.waitUntilAttributeExists('html', 'data-busy', 5000);

// wait until a HTML attribute no longer exists
helpers.waitUntilAttributeDoesNotExists('html', 'data-busy', 5000);

// get the content value of a :before pseudo element
helpers.getPseudoElementBeforeValue('body header');

// get the content value of a :after pseudo element
helpers.getPseudoElementAfterValue('body header');

// clear the cookies
helpers.clearCookies();

// clear both local and session storages
helpers.clearStorages();

// clear both cookies and storages
helpers.clearCookiesAndStorages('body header');

// waits until an element to exist and returns it
helpers.waitForCssXpathElement('#login-button', 5000);

// scroll until element is visible
helpers.scrollToElement(webElement);

// select a value inside a dropdown list by its text
helpers.selectByVisibleText('#country', 'Brazil');

// waits and returns an array of all windows opened
helpers.waitForNewWindows();

Visual Comparison

The selenium-cucumber-js framework uses Applitools Eyes to add visual checkpoints to your JavaScript Selenium tests. It takes care of getting screenshots of your application from the underlying WebDriver, sending them to the Applitools Eyes server for validation and failing the test when differences are detected. To perform visual comparisons within your tests, obtain an Applitools Eyes API Key and assign it to the eye_key property of the selenium-cucumber-js.json config file in the root of your project.

For example the following configuration could be used with an increased timeout which allows enough time for visual checks:

{
  "eye_key": "Your_Api_Key",
  "timeout": 50000
}

And its usage within page Objects:

module.exports = {

    url: 'https://applitools.com/helloworld',

    elements: {
        clickme: by.tagName('button'),
        searchResultLink: by.css('div.g > h3 > a')
    },

    applitools_Eyes_Example: function () {

        // Start the test and set the browser's viewport size to 800x600.
        eyes.open(driver, 'Hello World!', 'My first Javascript test!',
            {width: 800, height: 600});

        // Navigate the browser to the "hello world!" web-site.
        driver.get(page.HelloWorld.elements.url);

        // Visual checkpoint #1.
        eyes.checkWindow('Main Page');

        // Click the "Click me!" button.
        driver.findElement(page.HelloWorld.elements.clickme).click();

        // Visual checkpoint #2.
        eyes.checkWindow('Click!');

        // End the test.
        eyes.close();
    }
};

Before/After hooks

You can register before and after handlers for features and scenarios:

module.exports = function () {

    // add a before feature hook
    this.BeforeFeature(function(feature, done) {
        console.log('BeforeFeature: ' + feature.getName());
        done();
    });

    // add an after feature hook
    this.AfterFeature(function(feature, done) {
        console.log('AfterFeature: ' + feature.getName());
        done();
    });

    // add before scenario hook
    this.BeforeScenario(function(scenario, done) {
        console.log('BeforeScenario: ' + scenario.getName());
        done();
    });

    // add after scenario hook
    this.AfterScenario(function(scenario, done) {
        console.log('AfterScenario: ' + scenario.getName());
        done();
    });
};

Reports

HTML and JSON reports are automatically generated and stored in the default ./reports folder. This location can be changed by providing a new path using the -r command line switch:

How to debug

Most selenium methods return a JavaScript Promise that is resolved when the method completes. The easiest way to step in with a debugger is to add a .then method to a selenium function and place a debugger statement within it, for example:

module.exports = function () {

    this.When(/^I search Google for "([^"]*)"$/, function (searchQuery, done) {

        driver.findElement(by.name('q')).then(function(input) {
            expect(input).to.exist;
            debugger; // <<- your IDE should step in at this point, with the browser open
            return input;
        })
        .then(function(input){
            input.sendKeys(searchQuery);
            input.sendKeys(selenium.Key.ENTER);

            done(); // <<- let cucumber know you're done
        });
    });
};

Directory structure

You can use the framework without any command line arguments if your application uses the following folder structure:

.
├── features
│   └── google-search.feature
├── step-definitions
│   └── google-search-steps.js
├── page-objects
│   └── google-search.js
└── shared-objects
│   ├── test-data.js
│   └── stuff.json
└── reports
    ├── cucumber-report.json
    └── cucumber-report.html

Demo

This project includes an example to help you get started. You can run the example using the following command:

node ./node_modules/selenium-cucumber-js/index.js

Bugs

Please raise bugs via the selenium-cucumber-js issue tracker and, if possible, please provide enough information to allow the bug to be reproduced.

Contributing

Everyone is very welcome to contribute to this project. You can contribute just by submitting bugs or suggesting improvements by opening an issue on GitHub.

Troubleshooting

IntelliJ Cucumber Plugin

IntelliJ based IDE's have a plugin that allows the tester to control click on a Given, When, Then statement within a Cucumber feature file and have the user taken to the associated step definition. This plugin relies on your project having the following folder structure:

.
└── features
   │   google-search.feature
   └── step_definitions
   │   └── google-search-steps.js
   └── page_objects
   │   └── google-search.js
   └── shared_objects
   │   ├── test-data.js
   │   └── stuff.json
   └── reports
       ├── cucumber-report.json
       └── cucumber-report.html

This can be achieved by restructuring your project to match the layout above (notice the underscores), and running your tests with the following switches:

node ./node_modules/selenium-cucumber-js/index.js -s ./features/step_definitions -p ./features/page_objects -o ./features/shared_objects -r ./features/reports

VSCode Cucumber Plugin

Visual Studio Code has also an extension for Cucumber (Gherkin) Language Support + Format + Steps/PageObjects Autocomplete. You can find how to install and use at Cucumber (Gherkin) Full Support.

Following the default structure, the settings.json should look like this:

{
    "cucumberautocomplete.steps": [
        "step-definitions/*.js"
    ],
    "cucumberautocomplete.syncfeatures": "features/*.feature",
    "cucumberautocomplete.strictGherkinCompletion": false,
    "cucumberautocomplete.onTypeFormat": true,
    "editor.quickSuggestions": {
        "comments": false,
        "strings": true,
        "other": true
    },
    "cucumberautocomplete.gherkinDefinitionPart": "(Given|When|Then)\\(",
}

License

Licensed under ISC License © John Doherty