jest-process-manager

This project was forked because the package jest-dev-server is no longer maintained.

Starts a server before your Jest tests and tears it down after.

Why

jest-playwright-preset or jest-puppeteer works great for running tests in Jest using your preferred end-to-end testing library. It's also useful for starting a local development server during the tests without letting Jest hang. This package extracts just the local development server spawning without any ties to Puppeteer.

Install

1npm install --save-dev jest-process-manager

Usage

jest-process-manager exports setup,teardown and getServers functions.

1// global-setup.js
2const { setup: setupDevServer } = require('jest-process-manager')
3
4module.exports = async function globalSetup() {
5  await setupDevServer({
6    command: `node config/start.js --port=3000`,
7    launchTimeout: 50000,
8    port: 3000,
9  })
10  // Your global setup
11}

It is also possible to specify several servers:

1// global-setup.js
2const { setup: setupDevServer } = require('jest-process-manager')
3
4module.exports = async function globalSetup() {
5  await setupDevServer([
6    {
7      command: 'node server.js',
8      port: 4444,
9    },
10    {
11      command: 'node server2.js',
12      port: 4445,
13    },
14  ])
15  // Your global setup
16}

1// global-setup.js
2const { setup: setupDevServer, getServers } = require('jest-process-manager')
3
4module.exports = async function globalSetup() {
5  await setupDevServer({
6    command: `node config/start.js --port=3000`,
7    launchTimeout: 50000,
8    port: 3000,
9  })
10  getServers.then(servers => {
11    // You can get to the servers and do whatever you want
12  })
13  // Your global setup
14}

1// global-teardown.js
2const { teardown: teardownDevServer } = require('jest-process-manager')
3
4module.exports = async function globalTeardown() {
5  await teardownDevServer()
6  // Your global teardown
7}

Options

command

Type: string, required.

Command to execute to start the port. Directly passed to spawnd.

1module.exports = {
2  command: 'npm run start',
3}

debug

Type: boolean, default to false.

Log server output, useful if server is crashing at start.

1module.exports = {
2  command: 'npm run start',
3  debug: true,
4}

launchTimeout

Type: number, default to 5000.

How many milliseconds to wait for the spawned server to be available before giving up. Defaults to wait-port's default.

1module.exports = {
2  command: 'npm run start',
3  launchTimeout: 30000,
4}

---

Following options are linked to spawnd.

host

Type: string, default to localhost.

Host to wait for activity on before considering the server running. Must be used in conjunction with port.

1module.exports = {
2  command: 'npm run start --port 3000',
3  host: 'customhost.com',
4  port: 3000,
5}

protocol

Type: (https, http, http-get, https-get, tcp, socket) default to tcp.

To wait for an HTTP or TCP endpoint before considering the server running, include http or tcp as a protocol. Must be used in conjunction with port. This give you ability to define resource prefix for wait-on package.

1module.exports = {
2  command: 'npm run start --port 3000',
3  protocol: 'http',
4  port: 3000,
5}

port

Type: number, default to 3000.

Port to wait for activity on before considering the server running. If not provided, the server is assumed to immediately be running.

1module.exports = {
2  command: 'npm run start --port 3000',
3  port: 3000,
4}

basePath

Type: string

Option for a basePath where server is running.

1module.exports = {
2  command: 'npm run start',
3  basePath: '/myservice',
4}

usedPortAction

Type: string (ask, error, ignore, kill) default to ask.

It defines the action to take if port is already used:

• ask: a prompt is shown to decide if you want to kill the process or not
• error: an errow is thrown
• ignore: your test are executed, we assume that the server is already started
• kill: the process is automatically killed without a prompt

1module.exports = {
2  command: 'npm run start --port 3000',
3  port: 3000,
4  usedPortAction: 'kill',
5}

waitOnScheme

jest-dev-server use the wait-on npm package to wait for resources to become available before calling callback.

Type: object, default to {}.

• delay: optional initial delay in ms, default 0
• interval: optional poll resource interval in ms, default 250ms
• log: optional flag which outputs to stdout, remaining resources waited on and when complete or errored
• reverse: optional flag to reverse operation so checks are for resources being NOT available, default false
• timeout: optional timeout in ms, default Infinity. Aborts with error
• tcpTimeout: optional tcp timeout in ms, default 300ms
• verbose: optional flag which outputs debug output, default false
• window: optional stabilization time in ms, default 750ms. Waits this amount of time for file sizes to stabilize or other resource availability to remain unchanged

Note: http(s) specific options, see https://github.com/request/request#readme for specific details

1module.exports = {
2  command: 'npm run start --port 3000',
3  port: 3000,
4  usedPortAction: 'kill',
5  waitOnScheme: {
6    delay: 1000,
7  },
8}

options

Options which will be passed down to the spawn of the process. For example environment variables:

1// global-setup.js
2const { setup: setupDevServer, getServers } = require('jest-process-manager')
3
4module.exports = async function globalSetup() {
5  await setupDevServer({
6    command: `node config/start.js --port=3000`,
7    launchTimeout: 50000,
8    port: 3000,
9    options: {
10      env: {
11        "FOO": "bar",
12      }
13    }
14  })
15  getServers.then(servers => {
16    // You can get to the servers and do whatever you want
17  })
18  // Your global setup
19}

Troubleshooting

• If using port makes the terminal to ask for root password although the port is valid and accessible then use usePortAction: 'ignore'.

License

https://github.com/playwright-community/jest-process-manager/blob/master/LICENSE