Test runner

The node:test module facilitates the creation of JavaScript tests. To access it:

import test from 'node:test';

This module is only available under the node: scheme.

Tests created via the test module consist of a single function that is processed in one of three ways:

1. A synchronous function that is considered failing if it throws an exception, and is considered passing otherwise.
2. A function that returns a Promise that is considered failing if the Promise rejects, and is considered passing if the Promise fulfills.
3. A function that receives a callback function. If the callback receives any truthy value as its first argument, the test is considered failing. If a falsy value is passed as the first argument to the callback, the test is considered passing. If the test function receives a callback function and also returns a Promise, the test will fail.

The following example illustrates how tests are written using the test module.

test('synchronous passing test', (t) => {
  // This test passes because it does not throw an exception.
  assert.strictEqual(1, 1);
});

test('synchronous failing test', (t) => {
  // This test fails because it throws an exception.
  assert.strictEqual(1, 2);
});

test('asynchronous passing test', async (t) => {
  // This test passes because the Promise returned by the async
  // function is settled and not rejected.
  assert.strictEqual(1, 1);
});

test('asynchronous failing test', async (t) => {
  // This test fails because the Promise returned by the async
  // function is rejected.
  assert.strictEqual(1, 2);
});

test('failing test using Promises', (t) => {
  // Promises can be used directly as well.
  return new Promise((resolve, reject) => {
    setImmediate(() => {
      reject(new Error('this will cause the test to fail'));
    });
  });
});

test('callback passing test', (t, done) => {
  // done() is the callback function. When the setImmediate() runs, it invokes
  // done() with no arguments.
  setImmediate(done);
});

test('callback failing test', (t, done) => {
  // When the setImmediate() runs, done() is invoked with an Error object and
  // the test fails.
  setImmediate(() => {
    done(new Error('callback failure'));
  });
});

If any tests fail, the process exit code is set to 1.

The test context's test() method allows subtests to be created. It allows you to structure your tests in a hierarchical manner, where you can create nested tests within a larger test. This method behaves identically to the top level test() function. The following example demonstrates the creation of a top level test with two subtests.

test('top level test', async (t) => {
  await t.test('subtest 1', (t) => {
    assert.strictEqual(1, 1);
  });

  await t.test('subtest 2', (t) => {
    assert.strictEqual(2, 2);
  });
});

Note: beforeEach and afterEach hooks are triggered between each subtest execution.

In this example, await is used to ensure that both subtests have completed. This is necessary because tests do not wait for their subtests to complete, unlike tests created within suites. Any subtests that are still outstanding when their parent finishes are cancelled and treated as failures. Any subtest failures cause the parent test to fail.

Individual tests can be skipped by passing the skip option to the test, or by calling the test context's skip() method as shown in the following example.

// The skip option is used, but no message is provided.
test('skip option', { skip: true }, (t) => {
  // This code is never executed.
});

// The skip option is used, and a message is provided.
test('skip option with message', { skip: 'this is skipped' }, (t) => {
  // This code is never executed.
});

test('skip() method', (t) => {
  // Make sure to return here as well if the test contains additional logic.
  t.skip();
});

test('skip() method with message', (t) => {
  // Make sure to return here as well if the test contains additional logic.
  t.skip('this is skipped');
});

Individual tests can be marked as flaky or incomplete by passing the todo option to the test, or by calling the test context's todo() method, as shown in the following example. These tests represent a pending implementation or bug that needs to be fixed. TODO tests are executed, but are not treated as test failures, and therefore do not affect the process exit code. If a test is marked as both TODO and skipped, the TODO option is ignored.

// The todo option is used, but no message is provided.
test('todo option', { todo: true }, (t) => {
  // This code is executed, but not treated as a failure.
  throw new Error('this does not fail the test');
});

// The todo option is used, and a message is provided.
test('todo option with message', { todo: 'this is a todo test' }, (t) => {
  // This code is executed.
});

test('todo() method', (t) => {
  t.todo();
});

test('todo() method with message', (t) => {
  t.todo('this is a todo test and is not treated as a failure');
  throw new Error('this does not fail the test');
});

Suites and tests can also be written using the describe() and it() functions. describe() is an alias for suite(), and it() is an alias for test().

describe('A thing', () => {
  it('should work', () => {
    assert.strictEqual(1, 1);
  });

  it('should be ok', () => {
    assert.strictEqual(2, 2);
  });

  describe('a nested thing', () => {
    it('should work', () => {
      assert.strictEqual(3, 3);
    });
  });
});

describe() and it() are imported from the node:test module.

import { describe, it } from 'node:test';

If Node.js is started with the --test-only command-line option, or test isolation is disabled, it is possible to skip all tests except for a selected subset by passing the only option to the tests that should run. When a test with the only option is set, all subtests are also run. If a suite has the only option set, all tests within the suite are run, unless it has descendants with the only option set, in which case only those tests are run.

When using subtests within a test()/it(), it is required to mark all ancestor tests with the only option to run only a selected subset of tests.

The test context's runOnly() method can be used to implement the same behavior at the subtest level. Tests that are not executed are omitted from the test runner output.

// Assume Node.js is run with the --test-only command-line option.
// The suite's 'only' option is set, so these tests are run.
test('this test is run', { only: true }, async (t) => {
  // Within this test, all subtests are run by default.
  await t.test('running subtest');

  // The test context can be updated to run subtests with the 'only' option.
  t.runOnly(true);
  await t.test('this subtest is now skipped');
  await t.test('this subtest is run', { only: true });

  // Switch the context back to execute all tests.
  t.runOnly(false);
  await t.test('this subtest is now run');

  // Explicitly do not run these tests.
  await t.test('skipped subtest 3', { only: false });
  await t.test('skipped subtest 4', { skip: true });
});

// The 'only' option is not set, so this test is skipped.
test('this test is not run', () => {
  // This code is not run.
  throw new Error('fail');
});

describe('a suite', () => {
  // The 'only' option is set, so this test is run.
  it('this test is run', { only: true }, () => {
    // This code is run.
  });

  it('this test is not run', () => {
    // This code is not run.
    throw new Error('fail');
  });
});

describe.only('a suite', () => {
  // The 'only' option is set, so this test is run.
  it('this test is run', () => {
    // This code is run.
  });

  it('this test is run', () => {
    // This code is run.
  });
});

The --test-name-pattern command-line option can be used to only run tests whose name matches the provided pattern, and the --test-skip-pattern option can be used to skip tests whose name matches the provided pattern. Test name patterns are interpreted as JavaScript regular expressions. The --test-name-pattern and --test-skip-pattern options can be specified multiple times in order to run nested tests. For each test that is executed, any corresponding test hooks, such as beforeEach(), are also run. Tests that are not executed are omitted from the test runner output.

Given the following test file, starting Node.js with the --test-name-pattern="test [1-3]" option would cause the test runner to execute test 1, test 2, and test 3. If test 1 did not match the test name pattern, then its subtests would not execute, despite matching the pattern. The same set of tests could also be executed by passing --test-name-pattern multiple times (e.g. --test-name-pattern="test 1", --test-name-pattern="test 2", etc.).

test('test 1', async (t) => {
  await t.test('test 2');
  await t.test('test 3');
});

test('Test 4', async (t) => {
  await t.test('Test 5');
  await t.test('test 6');
});

Test name patterns can also be specified using regular expression literals. This allows regular expression flags to be used. In the previous example, starting Node.js with --test-name-pattern="/test [4-5]/i" (or --test-skip-pattern="/test [4-5]/i") would match Test 4 and Test 5 because the pattern is case-insensitive.

To match a single test with a pattern, you can prefix it with all its ancestor test names separated by space, to ensure it is unique. For example, given the following test file:

describe('test 1', (t) => {
  it('some test');
});

describe('test 2', (t) => {
  it('some test');
});

Starting Node.js with --test-name-pattern="test 1 some test" would match only some test in test 1.

Test name patterns do not change the set of files that the test runner executes.

If both --test-name-pattern and --test-skip-pattern are supplied, tests must satisfy both requirements in order to be executed.

Once a test function finishes executing, the results are reported as quickly as possible while maintaining the order of the tests. However, it is possible for the test function to generate asynchronous activity that outlives the test itself. The test runner handles this type of activity, but does not delay the reporting of test results in order to accommodate it.

In the following example, a test completes with two setImmediate() operations still outstanding. The first setImmediate() attempts to create a new subtest. Because the parent test has already finished and output its results, the new subtest is immediately marked as failed, and reported later to the <TestsStream>.

The second setImmediate() creates an uncaughtException event. uncaughtException and unhandledRejection events originating from a completed test are marked as failed by the test module and reported as diagnostic warnings at the top level by the <TestsStream>.

test('a test that creates asynchronous activity', (t) => {
  setImmediate(() => {
    t.test('subtest that is created too late', (t) => {
      throw new Error('error1');
    });
  });

  setImmediate(() => {
    throw new Error('error2');
  });

  // The test finishes after this line.
});

The Node.js test runner supports running in watch mode by passing the --watch flag:

node --test --watch

In watch mode, the test runner will watch for changes to test files and their dependencies. When a change is detected, the test runner will rerun the tests affected by the change. The test runner will continue to run until the process is terminated.

The test runner supports specifying a module that will be evaluated before all tests are executed and can be used to setup global state or fixtures for tests. This is useful for preparing resources or setting up shared state that is required by multiple tests.

This module can export any of the following:

• A globalSetup function which runs once before all tests start
• A globalTeardown function which runs once after all tests complete

The module is specified using the --test-global-setup flag when running tests from the command line.

// setup-module.js
async function globalSetup() {
  // Setup shared resources, state, or environment
  console.log('Global setup executed');
  // Run servers, create files, prepare databases, etc.
}

async function globalTeardown() {
  // Clean up resources, state, or environment
  console.log('Global teardown executed');
  // Close servers, remove files, disconnect from databases, etc.
}

module.exports = { globalSetup, globalTeardown };

If the global setup function throws an error, no tests will be run and the process will exit with a non-zero exit code. The global teardown function will not be called in this case.

The Node.js test runner can be invoked from the command line by passing the --test flag:

node --test

By default, Node.js will run all files matching these patterns:

• **/*.test.{cjs,mjs,js}
• **/*-test.{cjs,mjs,js}
• **/*_test.{cjs,mjs,js}
• **/test-*.{cjs,mjs,js}
• **/test.{cjs,mjs,js}
• **/test/**/*.{cjs,mjs,js}

Unless --no-experimental-strip-types is supplied, the following additional patterns are also matched:

• **/*.test.{cts,mts,ts}
• **/*-test.{cts,mts,ts}
• **/*_test.{cts,mts,ts}
• **/test-*.{cts,mts,ts}
• **/test.{cts,mts,ts}
• **/test/**/*.{cts,mts,ts}

Alternatively, one or more glob patterns can be provided as the final argument(s) to the Node.js command, as shown below. Glob patterns follow the behavior of glob(7). The glob patterns should be enclosed in double quotes on the command line to prevent shell expansion, which can reduce portability across systems.

node --test "**/*.test.js" "**/*.spec.js"

Matching files are executed as test files. More information on the test file execution can be found in the test runner execution model section.

When process-level test isolation is enabled, each matching test file is executed in a separate child process. The maximum number of child processes running at any time is controlled by the --test-concurrency flag. If the child process finishes with an exit code of 0, the test is considered passing. Otherwise, the test is considered to be a failure. Test files must be executable by Node.js, but are not required to use the node:test module internally.

Each test file is executed as if it was a regular script. That is, if the test file itself uses node:test to define tests, all of those tests will be executed within a single application thread, regardless of the value of the concurrency option of test().

When process-level test isolation is disabled, each matching test file is imported into the test runner process. Once all test files have been loaded, the top level tests are executed with a concurrency of one. Because the test files are all run within the same context, it is possible for tests to interact with each other in ways that are not possible when isolation is enabled. For example, if a test relies on global state, it is possible for that state to be modified by a test originating from another file.

When Node.js is started with the --experimental-test-coverage command-line flag, code coverage is collected and statistics are reported once all tests have completed. If the NODE_V8_COVERAGE environment variable is used to specify a code coverage directory, the generated V8 coverage files are written to that directory. Node.js core modules and files within node_modules/ directories are, by default, not included in the coverage report. However, they can be explicitly included via the --test-coverage-include flag. By default all the matching test files are excluded from the coverage report. Exclusions can be overridden by using the --test-coverage-exclude flag. If coverage is enabled, the coverage report is sent to any test reporters via the 'test:coverage' event.

Coverage can be disabled on a series of lines using the following comment syntax:

/* node:coverage disable */
if (anAlwaysFalseCondition) {
  // Code in this branch will never be executed, but the lines are ignored for
  // coverage purposes. All lines following the 'disable' comment are ignored
  // until a corresponding 'enable' comment is encountered.
  console.log('this is never executed');
}
/* node:coverage enable */

Coverage can also be disabled for a specified number of lines. After the specified number of lines, coverage will be automatically reenabled. If the number of lines is not explicitly provided, a single line is ignored.

/* node:coverage ignore next */
if (anAlwaysFalseCondition) { console.log('this is never executed'); }

/* node:coverage ignore next 3 */
if (anAlwaysFalseCondition) {
  console.log('this is never executed');
}

The tap and spec reporters will print a summary of the coverage statistics. There is also an lcov reporter that will generate an lcov file which can be used as an in depth coverage report.

node --test --experimental-test-coverage --test-reporter=lcov --test-reporter-destination=lcov.info

• No test results are reported by this reporter.
• This reporter should ideally be used alongside another reporter.

The node:test module supports mocking during testing via a top-level mock object. The following example creates a spy on a function that adds two numbers together. The spy is then used to assert that the function was called as expected.

import assert from 'node:assert';
import { mock, test } from 'node:test';

test('spies on a function', () => {
  const sum = mock.fn((a, b) => {
    return a + b;
  });

  assert.strictEqual(sum.mock.callCount(), 0);
  assert.strictEqual(sum(3, 4), 7);
  assert.strictEqual(sum.mock.callCount(), 1);

  const call = sum.mock.calls[0];
  assert.deepStrictEqual(call.arguments, [3, 4]);
  assert.strictEqual(call.result, 7);
  assert.strictEqual(call.error, undefined);

  // Reset the globally tracked mocks.
  mock.reset();
});

The same mocking functionality is also exposed on the TestContext object of each test. The following example creates a spy on an object method using the API exposed on the TestContext. The benefit of mocking via the test context is that the test runner will automatically restore all mocked functionality once the test finishes.

test('spies on an object method', (t) => {
  const number = {
    value: 5,
    add(a) {
      return this.value + a;
    },
  };

  t.mock.method(number, 'add');
  assert.strictEqual(number.add.mock.callCount(), 0);
  assert.strictEqual(number.add(3), 8);
  assert.strictEqual(number.add.mock.callCount(), 1);

  const call = number.add.mock.calls[0];

  assert.deepStrictEqual(call.arguments, [3]);
  assert.strictEqual(call.result, 8);
  assert.strictEqual(call.target, undefined);
  assert.strictEqual(call.this, number);
});

Mocking timers is a technique commonly used in software testing to simulate and control the behavior of timers, such as setInterval and setTimeout, without actually waiting for the specified time intervals.

Refer to the MockTimers class for a full list of methods and features.

This allows developers to write more reliable and predictable tests for time-dependent functionality.

The example below shows how to mock setTimeout. Using .enable({ apis: ['setTimeout'] }); it will mock the setTimeout functions in the node:timers and node:timers/promises modules, as well as from the Node.js global context.

Note: Destructuring functions such as import { setTimeout } from 'node:timers' is currently not supported by this API.

import assert from 'node:assert';
import { mock, test } from 'node:test';

test('mocks setTimeout to be executed synchronously without having to actually wait for it', () => {
  const fn = mock.fn();

  // Optionally choose what to mock
  mock.timers.enable({ apis: ['setTimeout'] });
  setTimeout(fn, 9999);
  assert.strictEqual(fn.mock.callCount(), 0);

  // Advance in time
  mock.timers.tick(9999);
  assert.strictEqual(fn.mock.callCount(), 1);

  // Reset the globally tracked mocks.
  mock.timers.reset();

  // If you call reset mock instance, it will also reset timers instance
  mock.reset();
});

The same mocking functionality is also exposed in the mock property on the TestContext object of each test. The benefit of mocking via the test context is that the test runner will automatically restore all mocked timers functionality once the test finishes.

import assert from 'node:assert';
import { test } from 'node:test';

test('mocks setTimeout to be executed synchronously without having to actually wait for it', (context) => {
  const fn = context.mock.fn();

  // Optionally choose what to mock
  context.mock.timers.enable({ apis: ['setTimeout'] });
  setTimeout(fn, 9999);
  assert.strictEqual(fn.mock.callCount(), 0);

  // Advance in time
  context.mock.timers.tick(9999);
  assert.strictEqual(fn.mock.callCount(), 1);
});

The mock timers API also allows the mocking of the Date object. This is a useful feature for testing time-dependent functionality, or to simulate internal calendar functions such as Date.now().

The dates implementation is also part of the MockTimers class. Refer to it for a full list of methods and features.

Note: Dates and timers are dependent when mocked together. This means that if you have both the Date and setTimeout mocked, advancing the time will also advance the mocked date as they simulate a single internal clock.

The example below show how to mock the Date object and obtain the current Date.now() value.

import assert from 'node:assert';
import { test } from 'node:test';

test('mocks the Date object', (context) => {
  // Optionally choose what to mock
  context.mock.timers.enable({ apis: ['Date'] });
  // If not specified, the initial date will be based on 0 in the UNIX epoch
  assert.strictEqual(Date.now(), 0);

  // Advance in time will also advance the date
  context.mock.timers.tick(9999);
  assert.strictEqual(Date.now(), 9999);
});

If there is no initial epoch set, the initial date will be based on 0 in the Unix epoch. This is January 1st, 1970, 00:00:00 UTC. You can set an initial date by passing a now property to the .enable() method. This value will be used as the initial date for the mocked Date object. It can either be a positive integer, or another Date object.

import assert from 'node:assert';
import { test } from 'node:test';

test('mocks the Date object with initial time', (context) => {
  // Optionally choose what to mock
  context.mock.timers.enable({ apis: ['Date'], now: 100 });
  assert.strictEqual(Date.now(), 100);

  // Advance in time will also advance the date
  context.mock.timers.tick(200);
  assert.strictEqual(Date.now(), 300);
});

You can use the .setTime() method to manually move the mocked date to another time. This method only accepts a positive integer.

Note: This method will execute any mocked timers that are in the past from the new time.

In the below example we are setting a new time for the mocked date.

import assert from 'node:assert';
import { test } from 'node:test';

test('sets the time of a date object', (context) => {
  // Optionally choose what to mock
  context.mock.timers.enable({ apis: ['Date'], now: 100 });
  assert.strictEqual(Date.now(), 100);

  // Advance in time will also advance the date
  context.mock.timers.setTime(1000);
  context.mock.timers.tick(200);
  assert.strictEqual(Date.now(), 1200);
});

If you have any timer that's set to run in the past, it will be executed as if the .tick() method has been called. This is useful if you want to test time-dependent functionality that's already in the past.

import assert from 'node:assert';
import { test } from 'node:test';

test('runs timers as setTime passes ticks', (context) => {
  // Optionally choose what to mock
  context.mock.timers.enable({ apis: ['setTimeout', 'Date'] });
  const fn = context.mock.fn();
  setTimeout(fn, 1000);

  context.mock.timers.setTime(800);
  // Timer is not executed as the time is not yet reached
  assert.strictEqual(fn.mock.callCount(), 0);
  assert.strictEqual(Date.now(), 800);

  context.mock.timers.setTime(1200);
  // Timer is executed as the time is now reached
  assert.strictEqual(fn.mock.callCount(), 1);
  assert.strictEqual(Date.now(), 1200);
});

Using .runAll() will execute all timers that are currently in the queue. This will also advance the mocked date to the time of the last timer that was executed as if the time has passed.

import assert from 'node:assert';
import { test } from 'node:test';

test('runs timers as setTime passes ticks', (context) => {
  // Optionally choose what to mock
  context.mock.timers.enable({ apis: ['setTimeout', 'Date'] });
  const fn = context.mock.fn();
  setTimeout(fn, 1000);
  setTimeout(fn, 2000);
  setTimeout(fn, 3000);

  context.mock.timers.runAll();
  // All timers are executed as the time is now reached
  assert.strictEqual(fn.mock.callCount(), 3);
  assert.strictEqual(Date.now(), 3000);
});

Snapshot tests allow arbitrary values to be serialized into string values and compared against a set of known good values. The known good values are known as snapshots, and are stored in a snapshot file. Snapshot files are managed by the test runner, but are designed to be human readable to aid in debugging. Best practice is for snapshot files to be checked into source control along with your test files.

Snapshot files are generated by starting Node.js with the --test-update-snapshots command-line flag. A separate snapshot file is generated for each test file. By default, the snapshot file has the same name as the test file with a .snapshot file extension. This behavior can be configured using the snapshot.setResolveSnapshotPath() function. Each snapshot assertion corresponds to an export in the snapshot file.

An example snapshot test is shown below. The first time this test is executed, it will fail because the corresponding snapshot file does not exist.

// test.js
suite('suite of snapshot tests', () => {
  test('snapshot test', (t) => {
    t.assert.snapshot({ value1: 1, value2: 2 });
    t.assert.snapshot(5);
  });
});

Generate the snapshot file by running the test file with --test-update-snapshots. The test should pass, and a file named test.js.snapshot is created in the same directory as the test file. The contents of the snapshot file are shown below. Each snapshot is identified by the full name of test and a counter to differentiate between snapshots in the same test.

exports[`suite of snapshot tests > snapshot test 1`] = `
{
  "value1": 1,
  "value2": 2
}
`;

exports[`suite of snapshot tests > snapshot test 2`] = `
5
`;

Once the snapshot file is created, run the tests again without the --test-update-snapshots flag. The tests should pass now.

The node:test module supports passing --test-reporter flags for the test runner to use a specific reporter.

The following built-reporters are supported:

• spec The spec reporter outputs the test results in a human-readable format. This is the default reporter.

• tap The tap reporter outputs the test results in the TAP format.

• dot The dot reporter outputs the test results in a compact format, where each passing test is represented by a ., and each failing test is represented by a X.

• junit The junit reporter outputs test results in a jUnit XML format

• lcov The lcov reporter outputs test coverage when used with the --experimental-test-coverage flag.

The exact output of these reporters is subject to change between versions of Node.js, and should not be relied on programmatically. If programmatic access to the test runner's output is required, use the events emitted by the <TestsStream>.

The reporters are available via the node:test/reporters module:

import { tap, spec, dot, junit, lcov } from 'node:test/reporters';

--test-reporter can be used to specify a path to custom reporter. A custom reporter is a module that exports a value accepted by stream.compose. Reporters should transform events emitted by a <TestsStream>

Example of a custom reporter using <stream.Transform>:

import { Transform } from 'node:stream';

const customReporter = new Transform({
  writableObjectMode: true,
  transform(event, encoding, callback) {
    switch (event.type) {
      case 'test:dequeue':
        callback(null, `test ${event.data.name} dequeued`);
        break;
      case 'test:enqueue':
        callback(null, `test ${event.data.name} enqueued`);
        break;
      case 'test:watch:drained':
        callback(null, 'test watch queue drained');
        break;
      case 'test:watch:restarted':
        callback(null, 'test watch restarted due to file change');
        break;
      case 'test:start':
        callback(null, `test ${event.data.name} started`);
        break;
      case 'test:pass':
        callback(null, `test ${event.data.name} passed`);
        break;
      case 'test:fail':
        callback(null, `test ${event.data.name} failed`);
        break;
      case 'test:plan':
        callback(null, 'test plan');
        break;
      case 'test:diagnostic':
      case 'test:stderr':
      case 'test:stdout':
        callback(null, event.data.message);
        break;
      case 'test:coverage': {
        const { totalLineCount } = event.data.summary.totals;
        callback(null, `total line count: ${totalLineCount}\n`);
        break;
      }
    }
  },
});

export default customReporter;

Example of a custom reporter using a generator function:

export default async function * customReporter(source) {
  for await (const event of source) {
    switch (event.type) {
      case 'test:dequeue':
        yield `test ${event.data.name} dequeued\n`;
        break;
      case 'test:enqueue':
        yield `test ${event.data.name} enqueued\n`;
        break;
      case 'test:watch:drained':
        yield 'test watch queue drained\n';
        break;
      case 'test:watch:restarted':
        yield 'test watch restarted due to file change\n';
        break;
      case 'test:start':
        yield `test ${event.data.name} started\n`;
        break;
      case 'test:pass':
        yield `test ${event.data.name} passed\n`;
        break;
      case 'test:fail':
        yield `test ${event.data.name} failed\n`;
        break;
      case 'test:plan':
        yield 'test plan\n';
        break;
      case 'test:diagnostic':
      case 'test:stderr':
      case 'test:stdout':
        yield `${event.data.message}\n`;
        break;
      case 'test:coverage': {
        const { totalLineCount } = event.data.summary.totals;
        yield `total line count: ${totalLineCount}\n`;
        break;
      }
    }
  }
}

The value provided to --test-reporter should be a string like one used in an import() in JavaScript code, or a value provided for --import.

The --test-reporter flag can be specified multiple times to report test results in several formats. In this situation it is required to specify a destination for each reporter using --test-reporter-destination. Destination can be stdout, stderr, or a file path. Reporters and destinations are paired according to the order they were specified.

In the following example, the spec reporter will output to stdout, and the dot reporter will output to file.txt:

node --test-reporter=spec --test-reporter=dot --test-reporter-destination=stdout --test-reporter-destination=file.txt

When a single reporter is specified, the destination will default to stdout, unless a destination is explicitly provided.

Note: shard is used to horizontally parallelize test running across machines or processes, ideal for large-scale executions across varied environments. It's incompatible with watch mode, tailored for rapid code iteration by automatically rerunning tests on file changes.

import { tap } from 'node:test/reporters';
import { run } from 'node:test';
import process from 'node:process';
import path from 'node:path';

run({ files: [path.resolve('./tests/test.js')] })
 .on('test:fail', () => {
   process.exitCode = 1;
 })
 .compose(tap)
 .pipe(process.stdout);

The suite() function is imported from the node:test module.

suite.skip(name?, options?, fn?)

Shorthand for skipping a suite. This is the same as suite([name], { skip: true }[, fn]).

suite.todo(name?, options?, fn?)

Shorthand for marking a suite as TODO. This is the same as suite([name], { todo: true }[, fn]).

suite.only(name?, options?, fn?)

Shorthand for marking a suite as only. This is the same as suite([name], { only: true }[, fn]).

The test() function is the value imported from the test module. Each invocation of this function results in reporting the test to the <TestsStream>.

The TestContext object passed to the fn argument can be used to perform actions related to the current test. Examples include skipping the test, adding additional diagnostic information, or creating subtests.

test() returns a Promise that fulfills once the test completes. if test() is called within a suite, it fulfills immediately. The return value can usually be discarded for top level tests. However, the return value from subtests should be used to prevent the parent test from finishing first and cancelling the subtest as shown in the following example.

test('top level test', async (t) => {
  // The setTimeout() in the following subtest would cause it to outlive its
  // parent test if 'await' is removed on the next line. Once the parent test
  // completes, it will cancel any outstanding subtests.
  await t.test('longer running subtest', async (t) => {
    return new Promise((resolve, reject) => {
      setTimeout(resolve, 1000);
    });
  });
});

The timeout option can be used to fail the test if it takes longer than timeout milliseconds to complete. However, it is not a reliable mechanism for canceling tests because a running test might block the application thread and thus prevent the scheduled cancellation.

test.skip(name?, options?, fn?)

Shorthand for skipping a test, same as test([name], { skip: true }[, fn]).

test.todo(name?, options?, fn?)

Shorthand for marking a test as TODO, same as test([name], { todo: true }[, fn]).

test.only(name?, options?, fn?)

Shorthand for marking a test as only, same as test([name], { only: true }[, fn]).

Alias for suite().

The describe() function is imported from the node:test module.

describe.skip(name?, options?, fn?)

Shorthand for skipping a suite. This is the same as describe([name], { skip: true }[, fn]).

describe.todo(name?, options?, fn?)

Shorthand for marking a suite as TODO. This is the same as describe([name], { todo: true }[, fn]).

describe.only(name?, options?, fn?)

Shorthand for marking a suite as only. This is the same as describe([name], { only: true }[, fn]).

Alias for test().

The it() function is imported from the node:test module.

it.skip(name?, options?, fn?)

Shorthand for skipping a test, same as it([name], { skip: true }[, fn]).

it.todo(name?, options?, fn?)

Shorthand for marking a test as TODO, same as it([name], { todo: true }[, fn]).

it.only(name?, options?, fn?)

Shorthand for marking a test as only, same as it([name], { only: true }[, fn]).

This function creates a hook that runs before executing a suite.

describe('tests', async () => {
  before(() => console.log('about to run some test'));
  it('is a subtest', () => {
    assert.ok('some relevant assertion here');
  });
});

This function creates a hook that runs after executing a suite.

describe('tests', async () => {
  after(() => console.log('finished running tests'));
  it('is a subtest', () => {
    assert.ok('some relevant assertion here');
  });
});

Note: The after hook is guaranteed to run, even if tests within the suite fail.

This function creates a hook that runs before each test in the current suite.

describe('tests', async () => {
  beforeEach(() => console.log('about to run a test'));
  it('is a subtest', () => {
    assert.ok('some relevant assertion here');
  });
});

This function creates a hook that runs after each test in the current suite. The afterEach() hook is run even if the test fails.

describe('tests', async () => {
  afterEach(() => console.log('finished running a test'));
  it('is a subtest', () => {
    assert.ok('some relevant assertion here');
  });
});

An object whose methods are used to configure available assertions on the TestContext objects in the current process. The methods from node:assert and snapshot testing functions are available by default.

It is possible to apply the same configuration to all files by placing common configuration code in a module preloaded with --require or --import.

assert.register(name, fn)

Defines a new assertion function with the provided name and function. If an assertion already exists with the same name, it is overwritten.

An object whose methods are used to configure default snapshot settings in the current process. It is possible to apply the same configuration to all files by placing common configuration code in a module preloaded with --require or --import.

snapshot.setDefaultSnapshotSerializers(serializers)

This function is used to customize the default serialization mechanism used by the test runner. By default, the test runner performs serialization by calling JSON.stringify(value, null, 2) on the provided value. JSON.stringify() does have limitations regarding circular structures and supported data types. If a more robust serialization mechanism is required, this function should be used.

snapshot.setResolveSnapshotPath(fn)

This function is used to customize the location of the snapshot file used for snapshot testing. By default, the snapshot filename is the same as the entry point filename with a .snapshot file extension.

The MockFunctionContext class is used to inspect or manipulate the behavior of mocks created via the MockTracker APIs.

A getter that returns a copy of the internal array used to track calls to the mock. Each entry in the array is an object with the following properties.

ctx.callCount(): integer

This function returns the number of times that this mock has been invoked. This function is more efficient than checking ctx.calls.length because ctx.calls is a getter that creates a copy of the internal call tracking array.

ctx.mockImplementation(implementation)

This function is used to change the behavior of an existing mock.

The following example creates a mock function using t.mock.fn(), calls the mock function, and then changes the mock implementation to a different function.

test('changes a mock behavior', (t) => {
  let cnt = 0;

  function addOne() {
    cnt++;
    return cnt;
  }

  function addTwo() {
    cnt += 2;
    return cnt;
  }

  const fn = t.mock.fn(addOne);

  assert.strictEqual(fn(), 1);
  fn.mock.mockImplementation(addTwo);
  assert.strictEqual(fn(), 3);
  assert.strictEqual(fn(), 5);
});

ctx.mockImplementationOnce(implementation, onCall?)

This function is used to change the behavior of an existing mock for a single invocation. Once invocation onCall has occurred, the mock will revert to whatever behavior it would have used had mockImplementationOnce() not been called.

The following example creates a mock function using t.mock.fn(), calls the mock function, changes the mock implementation to a different function for the next invocation, and then resumes its previous behavior.

test('changes a mock behavior once', (t) => {
  let cnt = 0;

  function addOne() {
    cnt++;
    return cnt;
  }

  function addTwo() {
    cnt += 2;
    return cnt;
  }

  const fn = t.mock.fn(addOne);

  assert.strictEqual(fn(), 1);
  fn.mock.mockImplementationOnce(addTwo);
  assert.strictEqual(fn(), 3);
  assert.strictEqual(fn(), 4);
});

ctx.resetCalls()

Resets the call history of the mock function.

ctx.restore()

Resets the implementation of the mock function to its original behavior. The mock can still be used after calling this function.

The MockModuleContext class is used to manipulate the behavior of module mocks created via the MockTracker APIs.

ctx.restore()

Resets the implementation of the mock module.

The MockPropertyContext class is used to inspect or manipulate the behavior of property mocks created via the MockTracker APIs.

A getter that returns a copy of the internal array used to track accesses (get/set) to the mocked property. Each entry in the array is an object with the following properties:

ctx.accessCount(): integer

This function returns the number of times that the property was accessed. This function is more efficient than checking ctx.accesses.length because ctx.accesses is a getter that creates a copy of the internal access tracking array.

ctx.mockImplementation(value)

This function is used to change the value returned by the mocked property getter.

ctx.mockImplementationOnce(value, onAccess?)

This function is used to change the behavior of an existing mock for a single invocation. Once invocation onAccess has occurred, the mock will revert to whatever behavior it would have used had mockImplementationOnce() not been called.

The following example creates a mock function using t.mock.property(), calls the mock property, changes the mock implementation to a different value for the next invocation, and then resumes its previous behavior.

test('changes a mock behavior once', (t) => {
  const obj = { foo: 1 };

  const prop = t.mock.property(obj, 'foo', 5);

  assert.strictEqual(obj.foo, 5);
  prop.mock.mockImplementationOnce(25);
  assert.strictEqual(obj.foo, 25);
  assert.strictEqual(obj.foo, 5);
});

For consistency with the rest of the mocking API, this function treats both property gets and sets as accesses. If a property set occurs at the same access index, the "once" value will be consumed by the set operation, and the mocked property value will be changed to the "once" value. This may lead to unexpected behavior if you intend the "once" value to only be used for a get operation.

ctx.resetAccesses()

Resets the access history of the mocked property.

ctx.restore()

Resets the implementation of the mock property to its original behavior. The mock can still be used after calling this function.

The MockTracker class is used to manage mocking functionality. The test runner module provides a top level mock export which is a MockTracker instance. Each test also provides its own MockTracker instance via the test context's mock property.

mock.fn(original?, implementation?, options?): Proxy

This function is used to create a mock function.

The following example creates a mock function that increments a counter by one on each invocation. The times option is used to modify the mock behavior such that the first two invocations add two to the counter instead of one.

test('mocks a counting function', (t) => {
  let cnt = 0;

  function addOne() {
    cnt++;
    return cnt;
  }

  function addTwo() {
    cnt += 2;
    return cnt;
  }

  const fn = t.mock.fn(addOne, addTwo, { times: 2 });

  assert.strictEqual(fn(), 2);
  assert.strictEqual(fn(), 4);
  assert.strictEqual(fn(), 5);
  assert.strictEqual(fn(), 6);
});

mock.getter(object, methodName, implementation?, options?)

This function is syntax sugar for MockTracker.method with options.getter set to true.

mock.method(object, methodName, implementation?, options?): Proxy

This function is used to create a mock on an existing object method. The following example demonstrates how a mock is created on an existing object method.

test('spies on an object method', (t) => {
  const number = {
    value: 5,
    subtract(a) {
      return this.value - a;
    },
  };

  t.mock.method(number, 'subtract');
  assert.strictEqual(number.subtract.mock.callCount(), 0);
  assert.strictEqual(number.subtract(3), 2);
  assert.strictEqual(number.subtract.mock.callCount(), 1);

  const call = number.subtract.mock.calls[0];

  assert.deepStrictEqual(call.arguments, [3]);
  assert.strictEqual(call.result, 2);
  assert.strictEqual(call.error, undefined);
  assert.strictEqual(call.target, undefined);
  assert.strictEqual(call.this, number);
});

mock.module(specifier, options?): MockModuleContext

This function is used to mock the exports of ECMAScript modules, CommonJS modules, JSON modules, and Node.js builtin modules. Any references to the original module prior to mocking are not impacted. In order to enable module mocking, Node.js must be started with the --experimental-test-module-mocks command-line flag.

The following example demonstrates how a mock is created for a module.

test('mocks a builtin module in both module systems', async (t) => {
  // Create a mock of 'node:readline' with a named export named 'fn', which
  // does not exist in the original 'node:readline' module.
  const mock = t.mock.module('node:readline', {
    namedExports: { fn() { return 42; } },
  });

  let esmImpl = await import('node:readline');
  let cjsImpl = require('node:readline');

  // cursorTo() is an export of the original 'node:readline' module.
  assert.strictEqual(esmImpl.cursorTo, undefined);
  assert.strictEqual(cjsImpl.cursorTo, undefined);
  assert.strictEqual(esmImpl.fn(), 42);
  assert.strictEqual(cjsImpl.fn(), 42);

  mock.restore();

  // The mock is restored, so the original builtin module is returned.
  esmImpl = await import('node:readline');
  cjsImpl = require('node:readline');

  assert.strictEqual(typeof esmImpl.cursorTo, 'function');
  assert.strictEqual(typeof cjsImpl.cursorTo, 'function');
  assert.strictEqual(esmImpl.fn, undefined);
  assert.strictEqual(cjsImpl.fn, undefined);
});

mock.property(object, propertyName, value?): Proxy

Creates a mock for a property value on an object. This allows you to track and control access to a specific property, including how many times it is read (getter) or written (setter), and to restore the original value after mocking.

test('mocks a property value', (t) => {
  const obj = { foo: 42 };
  const prop = t.mock.property(obj, 'foo', 100);

  assert.strictEqual(obj.foo, 100);
  assert.strictEqual(prop.mock.accessCount(), 1);
  assert.strictEqual(prop.mock.accesses[0].type, 'get');
  assert.strictEqual(prop.mock.accesses[0].value, 100);

  obj.foo = 200;
  assert.strictEqual(prop.mock.accessCount(), 2);
  assert.strictEqual(prop.mock.accesses[1].type, 'set');
  assert.strictEqual(prop.mock.accesses[1].value, 200);

  prop.mock.restore();
  assert.strictEqual(obj.foo, 42);
});