Automated Testing in the Pipeline

A pipeline that doesn't run your test suite isn't really "CI", it's just automated building. This lesson covers making tests a real, fast, reliable gate in your pipeline.

Fail fast

jobs:
  ci:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-node@v4
        with: { node-version: '20' }
      - run: npm ci
      - run: npm run lint
      - run: npm run typecheck
      - run: npm test

Steps run in order and stop at the first failure by default, if npm run lint fails, npm test never runs, no point spending minutes running a full test suite against code that doesn't even pass linting.

Caching dependencies

Reinstalling every dependency from scratch on every single run wastes minutes per build, adding up fast across a busy repository. Caching the package manager's download cache (not node_modules itself, which can hide subtle bugs if reused stale) speeds this up significantly:

- uses: actions/setup-node@v4
  with:
    node-version: '20'
    cache: 'npm'

setup-node's built-in cache: 'npm' option automatically caches and restores npm's download cache between runs, keyed on your lockfile, install still runs, but mostly from local cache instead of re-downloading everything.

Test matrices

To verify your code works across multiple versions or environments, define a matrix, GitHub Actions runs the job once per combination automatically:

jobs:
  test:
    runs-on: ubuntu-latest
    strategy:
      matrix:
        node-version: ['18', '20', '22']
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-node@v4
        with:
          node-version: ${{ matrix.node-version }}
      - run: npm ci && npm test

This runs the exact same job three times, once per Node version, in parallel, catching version-specific breakage before a user does.