Playwright Environment support

Installation

npm install @playwright/test @playwright-labs/fixture-env # npm
pnpm add @playwright/test @playwright-labs/fixture-env # pnpm
yarn add @playwright/test @playwright-labs/fixture-env # yarn

Define your schema (ajv-ts)

Install if needed

npm install ajv-ts # npm
pnpm add ajv-ts # pnpm
yarn add ajv-ts # yarn

// filename: env.ts
import { s } from "ajv-ts";

export const env = createEnv({
  prefix: "PW_",
  schema: {
    DATABASE_URL: s.string().format("uri"), // expect PW_DATABASE_URL
    OPEN_AI_API_KEY: s.string().min(1), // expect PW_OPEN_AI_API_KEY
  },
  env: process.env,
  onValidationError: (error) => {
    console.error("Validation error:", errors);
    process.exit(1);
  },
  extends: [],
});

Define your schema (zod)

Install if needed

npm install zod # npm
pnpm add zod # pnpm
yarn add zod # yarn

// filename: env.ts
import z from "zod";

export const env = createEnv({
  prefix: "PW_",
  schema: {
    DATABASE_URL: s.url(), // expect PW_DATABASE_URL
    OPEN_AI_API_KEY: s.string(), // expect PW_OPEN_AI_API_KEY
  },
  env: process.env,
  onValidationError: (error) => {
    console.error("Validation error:", errors);
    process.exit(1);
  },
  extends: [],
});

Usage

// filename: playwright.config.ts
import { env } from "./env";

export default defineConfig({
  use: {
    env: env,
  },
});

merging schemas

In case of merging different schemas, you can use extends option.

Example:

import { createEnv } from "@playwright-labs/fixture-env/ajv-ts";
const myBaseEnv = createEnv({
  env: {
    NODE_ENV: process.env.NODE_ENV,
  },
});

const extended = createEnv({
  extends: [myBaseEnv],
  env: {
    another: "qwerty",
  },
});

console.log("extended:", extended); // { NODE_ENV: 'development', another: 'qwerty' }

Predefined env variables for different runners

Each (ajv-ts and zod) schemas you can find predefined env.variables for different execution runtime.

| name | description | | ------ | -------------------------------- | | github | Github actions env variables 1 | | gitlab | Gitlab runner env variables 2 |

Fixture

Here is a list of fixtures that fixture-env exports for you

• useEnv(newEnv: Record<string, string>) - returns an object with all env variables. newEnv by default is process.env
• setEnv(env: Record<string, string>) - Sets environment variables for the current test and restores them after the test.
• getEnvValue(key: string, env?: Record<string, string>) - Gets a value from the environment. Default for env is process.env
• getEnvValueOrThrow- Gets a value from the environment, throwing if it doesn't exist.
• hasEnvKey(key: string) - Checks if the environment variable exists.
• getEnvKeysWithPrefix(prefix: string, env?: Record<string, string>) - Gets all keys from the environment that start with the given prefix.
• getEnvEntriesWithPrefix(prefix: string, env?: Record<string, string>) - Gets all environment entries (key-value pairs) that match a prefix.
• stripEnvPrefix(prefix: string, env?: Record<string, string>) - Removes the prefix from all keys in the environment.
• clearEnvKeys(keys: string[]) - Clears specific environment variables for the test.
• getEnvKeys(env?: Record<string, string>) - Gets all keys from the environment.
• snapshotEnv(env?: Record<string, string>) - Creates a snapshot of the environment variables for the current test.

Expectations

Here is a list of expectations that fixture-env exports for you

• expect(value: string).toBeInEnv() - expects that value is in process.env
• expect(value: string).toBeInEnvWithValue(value: string) - Asserts that the key exists in process.env with the specified value.
• expect(value: string).toHaveEnvKeysWithPrefix(prefix: string) - Asserts that the key exists in process.env with the specified prefix.
• expect(value: string).toMatchEnvPattern(pattern: RegExp) - Asserts that the key exists in process.env with a value that matches the given pattern.
• expect(value: string).toBeEnvUrl() - Asserts that the key exists in process.env with a value that is a valid URL.