1. testee
Automated cross-browser testing made easy.
testee
Package: testee
Created by: bitovi
Last modified: Mon, 27 Jun 2022 05:25:58 GMT
Version: 0.10.2
License: MIT
Downloads: 1,181
Repository: https://github.com/bitovi/testee

Install

npm install testee
yarn add testee

Testee

Greenkeeper badge
NPM Version
Build Status
Gitter Chat

Automated cross-browser testing made easy.

  • Supports any local browser that supports socket.io
  • Supports BrowserStack, Electron and PhantomJS
  • Supports Mocha, Jasmine and QUnit
  • Code coverage
  • Local and remote file testing

Installation

Node.js >= 4 is required. To install, type this at the command line:

 npm install testee

You can choose to install globally with -g if you wish, but it is recommended that you install per-project so that CI environments and all members of your team have instant access to the same dependencies.

Command Line Usage

On the command line, you have the following options available:

  • -h, --help: output usage information
  • -V, --version: output the version number
  • -b, --browsers [name]: A comma separated list of browsers you want to run (default: phantom)
  • -R, --root [path|URL]: The server root path or URL the files are relative to
  • -p, --port [port]: The port to run the server on (default: 3621)
  • -r, --reporter [name]: The name of the reporter to use (default: Dot)
  • --reporter-options [options]: The reporter specific options (separated by a comma)
  • -c, --config [file]: Use this JSON or JS configuration file (can be overridden by command line options)
  • --timeout [seconds]: The per test timeout (in seconds)
  • --delay [ms]: When running multiple tests, the time to wait for the browser to shut down before starting it with a new page.
  • -s, --server: Only run the server
  • --coverage: Enable code coverage

Examples

Test with one or multiple files:

 testee test.html
 testee test1.html test2.html

Save keystrokes with a base path/URL:

 testee test1.html test2.html --root=/var/www/app/
 testee test1.html test2.html --root=http://yourapp/

Specify browsers:

 testee test.html --browsers=firefox,safari

Use a configuration file:

 testee test.html --config=testee.json
 testee test.html --config=testee.js

Development Flow

During development it is nice to have tests run when files change. Testee is unique in that you do not need Testee while developing. Test pages can be served locally by something like http-server or live-server and opened/reloaded in the browser at will. This is the recommended way to iterate when using Testee. Testee is better used to test against multiple browsers as part of your build and/or release process, or alongside Git hooks for commits or pushes.

Programmatic Usage

For custom scripts (including gulp):

 const testee = require('testee');

const browsers = 'phantom';
const config = { reporter: 'spec' };
const files = ['test/test.html'];

testee.test(files, browsers, config)
  .then(function() {
    console.log('done!')
  });

Grunt Usage

See the Testee Grunt plugin grunt-testee for more information.

Configuration API

A simple, local browser config (JSON) example with mostly default values could look like:

 {
  "port": 3621,
  "root": "/var/www/app/",
  "reporter": "dot",
  "timeout": 120,
  "delay": 1000,
  "tunnel": {
    "type": "local"
  },
  "launch": {
    "type": "local"
  },
  "browsers": [
    "firefox",
    {
      "browser": "chrome",
      "args": [
        "--headless",
        "--disable-gpu",
        "--no-sandbox",
        "--remote-debugging-port=9222"
      ]
    }
  ]
}

BrowserStack hosts virtual machines running specific versions of web browsers and is extremely useful for cross-browser testing. It will require a username and password. An advanced config (JS) that runs your tests on an iPad Mini and Samsung Galaxy S3 emulator using BrowserStack in a CI environment (outputting XUnit logs) could look like:

 require('dotenv').config();

const pkg = require('./package.json');

module.exports = {
    reporter: 'XUnit',
    coverage: {
      dir: 'coverage/',
      reporters: ['text', 'html'],
      ignore: ['node_modules']
    }
    tunnel: {
      type: 'browserstack',
      key: process.env.BROWSERSTACK_KEY
    },
    launch: {
      type: 'browserstack',
      username: process.env.BROWSERSTACK_USER,
      password: process.env.BROWSERSTACK_KEY,
      version: 4
    },
    browsers: [
      { project:pkg.name, build:pkg.version, os:'ios', device:'iPad Mini', os_version:'6.0' },
      { project:pkg.name, build:pkg.version, os:'android', device:'Samsung Galaxy S III', os_version:'4.1' }
    ]
  }
};

General Settings

delay

Type: Number
Default value: 1000
The delay (in milliseconds) between multiple test pages within a single browser.

port

Type: Number
Default value: 3621
The port of the static fileserver used to serve the tests. This will also be used by Localhost tunneling services.

reporter

Type: String
Default value: 'dot'
See Mocha reporters.

root

Type: String
Default value: process.cwd()
The root path (or base URL) of the static fileserver used to serve the tests. Any test file will be relative to this path.

timeout

Type: Number
Default value: 120
The time (in seconds) to wait for a test page to report back before an error is thrown. This timeout might, for example, occurr when the given file doesn't exist, the browser didn't start, or the localhost tunnel wasn't running.

Browser Settings

Choose between locally installed browsers and those provided remotely by BrowserStack.

Note: Depending on your OS, the target browsers should not already be open/running before using this library. Electron and PhantomJS are the exceptions, as they can always be started multiple times.

browsers

Type: Array
Default value: ['phantom']
The browsers that will be used to run tests. For local browsers, use a browser name string (see Launchpad for more info). For a remote/BrowserStack browser, use a browser object. To pass command-line arguments and other browser options to Launchpad, use an object containing the browser name and optional array of arguments and options. This browser object will spawn a Chrome Headless browser:

 {
  "browser": "chrome",
  "args": [
    "--headless",
    "--disable-gpu",
    "--no-sandbox",
    "--remote-debugging-port=9222"
  ]
}

launch

Type: Object

launch.password

Type: String
Default value: undefined
Your BrowserStack API key.

launch.type

Type: String
Default value: 'local'
The test environment. Possible values are 'local' and 'browserstack'.

launch.username

Type: String
Default value: undefined
Your BrowserStack username.

launch.version

Type: String
Default value: undefined
The BrowserStack API version you'd like to use. The recommended value is 4.

Code Coverage

These options are used to instrument and report code coverage using Istanbul.

coverage

Type: Object

coverage.dir

Type: String
Default value: './coverage'
The directory where the coverage data should be written. text reports will be written to the console.

coverage.ignore

Type: Array
Default value: ['text']
A list of regex patterns that match files to be ignored by coverage instrumentation and reporting.

Note: When using the --coverage option, coverage.ignore will default to ['node_modules'].

coverage.reporters

Type: Array
Default value: []
The type of reporter(s) to use. Available reporters.

Note: babel-plugin-istanbul instruments ES2015/ES6 code automatically.

Localhost Tunneling

A localhost tunneling service makes your local system available to the outside world. This is great if you want to run tests on another system which can't easily reach your local machine. Such a service is necessary for giving BrowserStack workers an endpoint to communicate with.

tunnel

Type: Object
Default value: { type: 'local' }
See miner documentation for all available tunneling services and options.

Electron & PhantomJS

If you plan to use these, be sure to add electron and/or phantomjs-prebuilt to your list of devDependencies, or have them installed globally on your system.

Chrome Headless

To easily leverage Chrome Headless, you can add puppeteer to your project as a development dependency and the following browser option to your config's browsers array:

 const puppeteer = require('puppeteer')
process.env.LAUNCHPAD_CHROMIUM = puppeteer.executablePath()

module.exports = {
  "browser": "chromium",
  "args": [
    "--headless",
    "--disable-gpu",
    "--no-sandbox",
    "--remote-debugging-port=9222"
  ]
}

CI integration

Because different reporters are supported for the test result output, it is easy to obtain XUnit style XML files that integrate with CI servers like Jenkins. Just use the 'XUnit' reporter and write the output into a file. The following example writes the XML result of a Firefox test to "testresults.xml":

 testee test.html --browsers=firefox --reporter=XUnit > testresults.xml

See available reporters.

Note: If your CI platform uses Ubuntu to run builds, such as Travis CI, additional configuration is needed for FireFox to run properly. See bellow for more details.

Launching FireFox on Linux

The Linux version of FireFox uses D-Bus which is not preinstalled on all Linux distributions, including Ubuntu, this can cause errors when launching tests, heavily pollute Testee's debug logs with error messages, and make FireFox run slow.

To solve this problem, we recommend installing dbus-x11 via the apt package manager before the build executes.

Here is an example configuration for Travis CI:

 language: node_js
node_js:
  - "6"
addons:
  firefox: "latest-esr"
  # Install D-Bus here
  apt:
    packages:
      - "dbus-x11"
before_install:
  - "export DISPLAY=:99.0"
  - "sh -e /etc/init.d/xvfb start"

Capturing console.log and console.error

All calls to console.log and console.error in a test are tracked. To get the output during a test run, set the DEBUG environment variable to testee:console-log:

 DEBUG=testee:console-log testee test.html --browsers=canary

Debugging

Detailed debugging information can be enabled in any environment (command line, Grunt, programatically) by setting the DEBUG environment variable to testee:*:

 DEBUG=testee:* testee test.html --browsers=canary

Client side configuration

In most cases there is no need to change your actual test code.

One exception is when you load your testing library using an asynchronous client side loader like Steal or RequireJS because Testee won't know which library adapters to attach. In this case, you need to call Testee.init() manually once the test library is loaded:

 <script>
  define(['qunit'], function() {
    // Needs to check because it will only be available
    // when running the test with Testee
    if(window.Testee) {
      window.Testee.init();
    }

    QUnit.start();
  });
</script>

In some testing environments, reporting test progress via REST may work better than socket.io:

 <script>
  window.Testee = {
    provider: {
      type: 'rest'
    }
  };
</script>

Troubleshooting

Browser caching

On Safari, disable caching by choosing Develop -> Disable Caches and also Develop -> Empty Caches. Without these preferences, updated tests may not load in the browser.

"UnhandledPromiseRejectionWarning" messages

If you see a message which looks like the one below, there is most likely an incorrect assertion somewhere in your tests. Please refer to issue #136 for an example. If the warning is unrelated to your assertions, please open an issue.

UnhandledPromiseRejectionWarning: Unhandled promise rejection (rejection id: 1):
Error: Test `title` should be a "string" but "number" was given instead.

RELATED POST

10 Must-Know Windows Shortcuts That Will Save You Time

10 Must-Know Windows Shortcuts That Will Save You Time

Arrays vs Linked Lists: Which is Better for Memory Management in Data Structures?

Arrays vs Linked Lists: Which is Better for Memory Management in Data Structures?

Navigating AWS Networking: Essential Hacks for Smooth Operation

Navigating AWS Networking: Essential Hacks for Smooth Operation

Achieving Stunning Visuals with Unity's Global Illumination

Achieving Stunning Visuals with Unity's Global Illumination

Nim's Hidden Gems: Lesser-known Features for Writing Efficient Code

Nim's Hidden Gems: Lesser-known Features for Writing Efficient Code