Skip to content


Subversion checkout URL

You can clone with
Download ZIP
QUnit runner for nodejs
Branch: master

QUnit testing framework for nodejs.

Build Status


  • cli
  • testrunner api
  • test coverage via istanbul
  • tests inside of one testfile run synchronous, but every testfile runs parallel
  • tests from each file run in its own spawned node process
  • same API for client and server side code (original QUnit is used)
  • the simplest API of the world, especially for asynchronous testing
  • you can write tests in TDD or BDD style depending on your task and test type
  • you can run the same tests in browser if there is no dependencies to node
  • generators support


$ npm i qunit


The only exception

// Separate tests into modules.
// Use `QUnit` namespace, because `module` is reserved for node.
QUnit.module(name, lifecycle)


Command line

Read full cli api doc using "--help" or "-h":

$ qunit -h

$ qunit -c ./code.js -t ./tests.js

By default, code and dependencies are added to the global scope. To specify requiring them into a namespace object, prefix the path or module name with the variable name to be used for the namespace object, followed by a colon:

$ qunit -c code:./code.js -d utils:utilmodule -t ./time.js

via api

var testrunner = require("qunit");

// Defaults:
    // logging options
    log: {

        // log assertions overview
        assertions: true,

        // log expected and actual values for failed tests
        errors: true,

        // log tests overview
        tests: true,

        // log summary
        summary: true,

        // log global summary (all files)
        globalSummary: true,

        // log coverage
        coverage: true,

        // log global coverage (all files)
        globalCoverage: true,

        // log currently testing code file
        testing: true

    // run test coverage tool
    coverage: false,

    // define dependencies, which are required then before code
    deps: null,

    // define namespace your code will be attached to on global['your namespace']
    namespace: null,

    // max amount of ms child can be blocked, after that we assume running an infinite loop
    maxBlockDuration: 2000
// change any option for all tests globally
testrunner.options.optionName = value;

// or use setup function
    log: {
        summary: true

// one code and tests file{
    code: "/path/to/your/code.js",
    tests: "/path/to/your/tests.js"
}, callback);

// require code into a namespace object, rather than globally{
    code: {path: "/path/to/your/code.js", namespace: "code"},
    tests: "/path/to/your/tests.js"
}, callback);

// one code and multiple tests file{
    code: "/path/to/your/code.js",
    tests: ["/path/to/your/tests.js", "/path/to/your/tests1.js"]
}, callback);

// array of code and test files[
        code: "/path/to/your/code.js",
        tests: "/path/to/your/tests.js"
        code: "/path/to/your/code.js",
        tests: "/path/to/your/tests.js"
], callback);

// using testrunner callback{
    code: "/path/to/your/code.js",
    tests: "/path/to/your/tests.js"
}, function(err, report) {

// specify dependency{
    deps: "/path/to/your/dependency.js",
    code: "/path/to/your/code.js",
    tests: "/path/to/your/tests.js"
}, callback);

// dependencies can be modules or files{
    deps: "modulename",
    code: "/path/to/your/code.js",
    tests: "/path/to/your/tests.js"
}, callback);

// dependencies can required into a namespace object{
    deps: {path: "utilmodule", namespace: "utils"},
    code: "/path/to/your/code.js",
    tests: "/path/to/your/tests.js"
}, callback);

// specify multiple dependencies{
    deps: ["/path/to/your/dependency1.js", "/path/to/your/dependency2.js"],
    code: "/path/to/your/code.js",
    tests: "/path/to/your/tests.js"
}, callback);

Writing tests

QUnit API and code which have to be tested are already loaded and attached to the global context.

Some tests examples

test("a basic test example", function () {
    ok(true, "this test is fine");
    var value = "hello";
    equal("hello", value, "We expect value to be hello");

QUnit.module("Module A");

test("first test within module", 1, function () {
    ok(true, "a dummy");

test("second test within module", 2, function () {
    ok(true, "dummy 1 of 2");
    ok(true, "dummy 2 of 2");

QUnit.module("Module B", {
    setup: function () {
        // do some initial stuff before every test for this module
    teardown: function () {
        // do some stuff after every test for this module

test("some other test", function () {
    equal(true, false, "failing test");
    equal(true, true, "passing test");

QUnit.module("Module C", {
    setup: function() {
        // setup a shared environment for each test
        this.options = { test: 123 };

test("this test is using shared environment", 1, function () {
    deepEqual({ test: 123 }, this.options, "passing test");

test("this is an async test example", function () {
    setTimeout(function () {
        ok(true, "finished async test");
        strictEqual(true, true, "Strict equal assertion uses ===");
    }, 100);

Generators support

test("my async test with generators", function* () {
    var data = yield asyncFn();
    equal(data, {a: 1}, 'generators work');

Run tests

$ npm i
$ npm test


Code coverage via Istanbul.

To utilize, install istanbul and set option coverage: true or give a path where to store report coverage: {dir: "coverage/path"} or pass --cov parameter in the shell.

To specify the format of coverage report pass reporters array to the coverage options: coverage: {reporters: ['lcov', 'json']} (default)

Coverage calculations based on code and tests passed to node-qunit.

Something went wrong with that request. Please try again.