Utility for launching your applications on a per-test basis.

npm install @epic-web/app-launcher

// remix.launcher.ts import { defineLauncher } from '@epic-web/app-launcher' import getPort from 'get-port' export const launcher = defineLauncher({ async context() { return { port: await getPort(), } }, env({ context }) { return { PORT: context.port.toString(), } }, command() { return `npm start` }, url({ context }) { return `http://localhost:${context.port}` }, })

import { launcher } from './remix.launcher.js' test('renders the homepage', async ({ page }) => { await using app = await launcher.run() await page.goto(app.url.href) await expect(page.getByText("Welcome!")).toBeVisible() })

You can create a launcher by extending an existing launcher. Provide the launcher object you want to extend as the value of the extends property when creating a launcher:

import { remoteUtils } from 'msw/node' import { launcher } from './remix.launcher.js' export const launcher = defineLauncher({ extends: launcher, async env() { return { [remoteUtils.variableName]: remoteUtils.getContextId(), } }, })

Extending an existing launcher will inherit the following of its properties:

• env

• options <Object>
  • context <Function> A function that returns the context object. The context object is later shared across the other options.
  • env <Function> A function that returns an object representing environment variables to add to the spawned application process.
  • command <Function> A function that returns a string indicating the CLI command to run the application (e.g. npm start).
  • url <Function> A function that returns the resolved application URL. You can compose the URL based on your context or wait for the appProcess stdout/stderr to print that URL, then return it.

• port <number> A port number to await.
• options <Object>
  • timeout <number> The timeout duration (in ms). If the port is still vacant after this duration has passed, the promise rejects.
  • retryInterval <number> The interval (in ms) between individual retries.
  • maxRetries <number> The total number of retries before rejecting the returned promise.
• Returns: <Promise>

Returns a promise that resolves whenever a process starts running at the given port.

MIT