Integration testing patterns in Node.js

Introduction

Integration testing patterns help check if different parts of your app work well together. They find problems that unit tests might miss.

When you want to test how your database and server code work together.

When you need to verify that API endpoints return correct data.

When you want to check if multiple modules communicate properly.

When you want to catch bugs caused by interaction between components.

When you want to test real workflows instead of isolated functions.

Syntax

Node.js

describe('Feature or module', () => {
  beforeAll(async () => {
    // setup code like connecting to DB
  });

  afterAll(async () => {
    // cleanup code like closing DB connection
  });

  test('should do something', async () => {
    const result = await someIntegrationFunction();
    expect(result).toBe(expectedValue);
  });
});

describe groups related tests.

beforeAll and afterAll help prepare and clean up resources.

Examples

Tests an API endpoint to check if it returns a list of users with status 200.

Node.js

describe('User API', () => {
  test('GET /users returns list', async () => {
    const response = await request(app).get('/users');
    expect(response.status).toBe(200);
    expect(Array.isArray(response.body)).toBe(true);
  });
});

Sets up and tears down a database connection, then tests saving and retrieving data.

Node.js

describe('Database connection', () => {
  beforeAll(async () => {
    await db.connect();
  });

  afterAll(async () => {
    await db.disconnect();
  });

  test('should save and retrieve data', async () => {
    await db.save({ id: 1, name: 'Test' });
    const data = await db.find(1);
    expect(data.name).toBe('Test');
  });
});

Sample Program

This example creates a simple Express app with one endpoint. The integration test checks if the endpoint returns the expected user list.

Node.js

import request from 'supertest';
import express from 'express';

const app = express();
app.use(express.json());

const users = [{ id: 1, name: 'Alice' }];

app.get('/users', (req, res) => {
  res.json(users);
});

// Integration test
import { describe, test, expect } from '@jest/globals';

describe('User API integration test', () => {
  test('GET /users returns users list', async () => {
    const response = await request(app).get('/users');
    expect(response.status).toBe(200);
    expect(Array.isArray(response.body)).toBe(true);
    expect(response.body.length).toBe(1);
    expect(response.body[0].name).toBe('Alice');
  });
});

Important Notes

Use real or test databases to avoid messing with production data.

Keep tests isolated to avoid side effects between tests.

Run integration tests after unit tests for faster feedback.

Summary

Integration testing patterns help check if parts of your app work together correctly.

Use setup and cleanup hooks to manage resources like databases.

Test real workflows like API calls or database operations.

Practice

1. What is the main goal of integration testing in a Node.js application?

easy

A. To check the syntax of the code

B. To test individual functions in isolation

C. To verify that different parts of the app work together correctly

D. To measure the app's performance under load

Solution

1. Step 1: Understand integration testing purpose

   Integration testing focuses on checking if multiple parts of the app interact correctly, not just individual pieces.

2. Step 2: Compare with other testing types

   Syntax checking is done by linters, unit tests check single functions, and performance tests measure speed, so these are not integration tests.

3. Final Answer:

   To verify that different parts of the app work together correctly -> Option C

4. Quick Check:

   Integration testing = check parts working together [OK]

Hint: Integration tests check combined parts, not single functions [OK]

Common Mistakes:

• Confusing integration tests with unit tests
• Thinking integration tests check code syntax
• Mixing integration tests with performance testing

2. Which of the following is the correct way to define a setup hook in a Node.js integration test using Mocha?

easy

A. init(() => { /* setup code */ })

B. setup(() => { /* setup code */ })

C. start(() => { /* setup code */ })

D. beforeEach(() => { /* setup code */ })

Solution

1. Step 1: Recall Mocha setup hooks

   Mocha uses beforeEach and before hooks to run setup code before tests.

2. Step 2: Identify correct syntax

   Only beforeEach is a valid Mocha hook; setup, init, and start are not Mocha functions.

3. Final Answer:

   beforeEach(() => { /* setup code */ }) -> Option D

4. Quick Check:

   Mocha setup hook = beforeEach [OK]

Hint: Mocha uses beforeEach for setup, not setup or init [OK]

Common Mistakes:

• Using non-existent hook names like setup or init
• Confusing beforeEach with afterEach
• Forgetting arrow function syntax

3. Consider this Node.js integration test snippet using Jest and Supertest:

test('GET /users returns 200', async () => {
  const response = await request(app).get('/users');
  expect(response.statusCode).toBe(200);
});

What does this test check?

medium

A. That the /users endpoint returns a JSON array

B. That the /users endpoint returns a 200 status code

C. That the app throws an error on /users

D. That the /users endpoint is unreachable

Solution

1. Step 1: Analyze the test code

   The test sends a GET request to '/users' and checks the response status code.

2. Step 2: Understand the assertion

   The assertion expects the status code to be 200, meaning success.

3. Final Answer:

   That the /users endpoint returns a 200 status code -> Option B

4. Quick Check:

   Response status 200 means success [OK]

Hint: Check what the expect() function tests in the code [OK]

Common Mistakes:

• Assuming it checks response body instead of status
• Thinking it tests error or unreachable endpoint
• Ignoring async/await usage

4. You wrote this integration test using Jest:

beforeAll(() => {
  db.connect();
});

test('fetch data', async () => {
  const data = await db.getData();
  expect(data).toBeDefined();
});

afterAll(() => {
  db.disconnect();
});

But the test fails because the database connection is not ready. What is the likely problem?

medium

A. beforeAll should be async and await db.connect()

B. afterAll should run before the test

C. db.getData() should not be awaited

D. expect(data).toBeDefined() is incorrect syntax

Solution

1. Step 1: Check async setup in beforeAll

   Database connection is asynchronous, so beforeAll must be async and await db.connect() to ensure readiness.

2. Step 2: Verify other parts

   afterAll runs after tests, awaiting db.getData() is correct, and expect syntax is valid.

3. Final Answer:

   beforeAll should be async and await db.connect() -> Option A

4. Quick Check:

   Async setup needs await in beforeAll [OK]

Hint: Async setup needs await in beforeAll hook [OK]

Common Mistakes:

• Not making beforeAll async for async setup
• Misplacing afterAll before tests
• Ignoring await on async calls

5. You want to write an integration test for a Node.js app that creates a user via POST /users and then fetches it via GET /users/:id. Which pattern best ensures your test is reliable and isolated?

hard

A. Use setup hook to create a test user, then test GET /users/:id, and cleanup hook to delete the user

B. Run both POST and GET requests in the same test without cleanup

C. Only test GET /users/:id assuming user exists in database

D. Manually add user in database before tests and never delete

Solution

1. Step 1: Understand test isolation

   Tests should not depend on external data; setup creates needed data, cleanup removes it.

2. Step 2: Apply setup and cleanup hooks

   Creating a user in setup, testing GET, then deleting user in cleanup keeps tests reliable and repeatable.

3. Final Answer:

   Use setup hook to create a test user, then test GET /users/:id, and cleanup hook to delete the user -> Option A

4. Quick Check:

   Setup and cleanup keep tests isolated and reliable [OK]

Hint: Setup and cleanup hooks keep tests clean and repeatable [OK]

Common Mistakes:

• Not cleaning up test data causing flaky tests
• Assuming data exists without setup
• Combining unrelated tests without isolation