scholar-runner

scholar-runner

NPM command line module for running visual regression tests

Synopsis

This is the client test runner for usage with the Scholar application.

Installation

npm install scholar-runner --save-dev

Usage

scholar-runner --help

Usage: scholar-runner [options]

  Options:
    -h, --help                                      Output usage information
    -V, --version                                   Output the version number
    -b, --browser <option>                          Define test browser (defaults to "phantomjs")
    -r, --runner <option>                           Define the runner for the tests e.g. local, remote (defaults to "local")
    --compareLocally                                Define to run the comparison locally (defaults to false)
    --seleniumVersion <version>                     Optionally use a specific selenium version
    --verbose                                       Define selenium log level
    -s, --suite <suite>                             Define file to run (optional)
    -t, --type <key>                                Define subset of tests to run (optional)
    -c, --config <filePath>                         Define config file location (defaults to "config/scholar.js")
    -d, --directory <testDirectory>                 Define test files directory (defaults to "process.cwd()/test/")
    -o, --output <imageDirectory>                   Define directory to place screenshots (defaults to "process.cwd()/test_images/")
    -r, --testReportDirectory <testReportDirectory> Define directory to place report html (defaults to "process.cwd()/test_images/")
    --baselineDirectory <baselineDirectory>         Define baseline files directory (defaults to "process.cwd()/baselines/")

Setup

Config

The minimum configuration required is:

module.exports = {
  baseUrl: 'http://the-website-you-want-to-test.com',
  scholarUrl: 'http://your-scholar-instance.com'
};

Adding Global Cookies to Config

Add required cookies to your config file,

module.exports = {
  baseUrl: 'http://the-website-you-want-to-test.com',
  scholarUrl: 'http://your-scholar-instance.com',
  cookies: [{
      'name': 'example',
      'value': 'exampleValue',
      'domain': '.example.com'
  }]
};

Remote Sessions

To run your tests against a remote Selenium server (ie BrowserStack / Saucelabs) you will need to:

• Set the runner flag to remote
• Set the host (required) and port number (optional) in your config file.
• Set the user variable in config to your Selenium login (optional).
• Set the key variable in config to your automation key (optional).

For Example:

module.exports = {
      baseUrl: 'http://the-website-you-want-to-test.com',
      scholarUrl: 'http://your-scholar-instance.com',
      host: 'http://your-Selenium-server.com',
      port: '1337',
      user: process.env.USER || 'user-name',
      key: process.env.SECRET_KEY || 'my-secret-key'
};

Running on BrowserStack:

• Define the user variable as your BrowserStack user in config.
• Define the key variable as your BrowserStack automation key in config.
• If you want to test somewhere not internet accessible (localhost / internal domain etc) you will need to start up the browserstack local tunnel CLI. It's not included with this package, however the following should help get started:
  • wget https://www.browserstack.com/browserstack-local/BrowserStackLocal-`YOURENVNAME`-x64.zip && unzip BrowserStackLocal-YOURENVNAME-x64.zip
  • ./BrowserStackLocal -v -onlyAutomate -forcelocal $BROWSERSTACK_KEY &
• If you are required to run BrowserStack with a local/private connection add browserstack.local:true to your config file.

An Example browserstack config:

{
    baseUrl: 'http://the-website-you-want-to-test.com',
    scholarUrl: 'http://your-scholar-instance.com',
    host: 'hub.browserstack.com',
    user: 'browserstack-user-name',
    key: 'browserstack-automation-key',
    browserstack: {
        'browserstack.local' : true,
        ie: {
            browser: 'IE',
            browser_version: '11',
            os: 'Windows',
            os_version: '8.1'
        },
        edge: {
            browser: 'Edge',
            browser_version: '12',
            os: 'Windows',
            os_version: '10'
        },
        safari: {
            browser: 'Safari',
            browser_version: '9.1',
            os: 'OS X',
            os_version: 'El Capitan'
        },
        chrome: {
            browser: 'Chrome',
            browser_version: '47',
            os: 'Windows',
            os_version: '7'
        }
    }
}

Writing Specs

Your first spec file

Within your test directory (defaults to test/) add your spec file. Ensure the ending contains '-spec.js';

let desktopSpecs = [
    {
        name: 'desktop-test-1',
        selector: 'body',
        path: '/'
    },
    {
        name: 'alex-test',
        selector: 'body',
        path: '/'
    }
];

let otherSpecs = [
    {
        name: 'other-test-1',
        selector: 'body',
        path: '/'
    },
    {
        name: 'other-test-2',
        selector: 'body',
        path: '/'
    }
];

module.exports = {
    desktop: desktopSpecs,
    another: otherSpecs,
    all: desktopSpecs.concat(otherSpecs) // default if no --type is provided
};

Advanced Options

There are other possible spec options such as:

• name: String that will be the unique identifier within Scholar.
  • Example = "TestHeaderDesktop"
• selector: CSS Selector to target specific elements.
  • Example = ".main.header"
• url: Optional override of the global baseUrl.
  • Example = "alternative.domain.testing.com"
• path: URL path to target page.
  • Example = "/myTestPage"
• labels: An array allowing labelling / tagging of images for use within scholar
  • Example = ['homepage', 'carousel', 'hero']
• loadTimeout: Milliseconds to allow page to stabilise before taking a screenshot (Defaults to 2000).
  • Example = 5000
• setup: function that will be passed to the runner to allow events to happen after page loads (click on an accordion heading etc). MUST return the client
  • Example = function(client) { return client.waitForVisible('.n-button', 5000).click('.n-button'); }
• setupTimeout: Milliseconds to allow page to stabilise after running a setup function (Defaults to 0).
  • Example = 2000
• waitTimeout: Milliseconds to wait for required element to become visible (Defaults to 3000).
  • Example = 5000
• viewportSize: JS Object with width and height properties.
  • Example = {width: 1280, height: 720}
• cookies: Same format as the config options, however will only be included for the specific test.
  • Example = [ { name: 'first', value: 'test' }, { name: 'second', value: 'another' } ]

License

GNU