Using environment variables

本頁內容尚未翻譯。

Astro gives you access to Vite’s built-in environment variables support and includes some default environment variables for your project that allow you to access configuration values for your current project (e.g. site, base), whether your project is running in development or production, and more.

Astro also provides a way to use and organize your environment variables with type safety. It is available for use inside the Astro context (e.g. Astro components, routes and endpoints, UI framework components, middleware), and managed with a schema in your Astro configuration.

Vite’s built-in support

Astro uses Vite’s built-in support for environment variables, which are statically replaced at build time, and lets you use any of its methods to work with them.

Note that while all environment variables are available in server-side code, only environment variables prefixed with PUBLIC_ are available in client-side code for security purposes.

SECRET_PASSWORD=password123
PUBLIC_ANYBODY=there

In this example, PUBLIC_ANYBODY (accessible via import.meta.env.PUBLIC_ANYBODY) will be available in server or client code, while SECRET_PASSWORD (accessible via import.meta.env.SECRET_PASSWORD) will be server-side only.

IntelliSense for TypeScript

By default, Astro provides a type definition for import.meta.env in astro/client.d.ts.

While you can define more custom env variables in .env.[mode] files, you may want to get TypeScript IntelliSense for user-defined env variables which are prefixed with PUBLIC_.

To achieve this, you can create an env.d.ts in src/ and configure ImportMetaEnv like this:

interface ImportMetaEnv {
  readonly DB_PASSWORD: string;
  readonly PUBLIC_POKEAPI: string;
  // more env variables...
}

interface ImportMeta {
  readonly env: ImportMetaEnv;
}

Default environment variables

Astro includes a few environment variables out of the box:

import.meta.env.MODE: The mode your site is running in. This is development when running astro dev and production when running astro build.
import.meta.env.PROD: true if your site is running in production; false otherwise.
import.meta.env.DEV: true if your site is running in development; false otherwise. Always the opposite of import.meta.env.PROD.
import.meta.env.BASE_URL: The base URL your site is being served from. This is determined by the base config option.
import.meta.env.SITE: This is set to the site option specified in your project’s astro.config.
import.meta.env.ASSETS_PREFIX: The prefix for Astro-generated asset links if the build.assetsPrefix config option is set. This can be used to create asset links not handled by Astro.

Use them like any other environment variable.

const isProd = import.meta.env.PROD;
const isDev = import.meta.env.DEV;

Setting environment variables

`.env` files

Environment variables can be loaded from .env files in your project directory.

Just create a .env file in the project directory and add some variables to it.

# This will only be available when run on the server!
DB_PASSWORD="foobar"
# This will be available everywhere!
PUBLIC_POKEAPI="https://pokeapi.co/api/v2"

You can also add .production, .development or a custom mode name to the filename itself (e.g env.testing, .env.staging). This allows you to use different sets of environment variables at different times.

The astro dev and astro build commands default to "development" and "production" modes, respectively. You can run these commands with the --mode flag to pass a different value for mode and load the matching .env file.

This allows you to run the dev server or build your site connecting to different APIs:

# Run the dev server connected to a "staging" API
astro dev --mode staging

# Build a site that connects to a "production" API with additional debug information
astro build --devOutput

# Build a site that connects to a "testing" API
astro build --mode testing

For more on .env files, see the Vite documentation.

Using the CLI

You can also add environment variables as you run your project:

PUBLIC_POKEAPI=https://pokeapi.co/api/v2 npm run dev

PUBLIC_POKEAPI=https://pokeapi.co/api/v2 pnpm run dev

PUBLIC_POKEAPI=https://pokeapi.co/api/v2 yarn run dev

Getting environment variables

Environment variables in Astro are accessed with import.meta.env, using the import.meta feature added in ES2020, instead of process.env.

For example, use import.meta.env.PUBLIC_POKEAPI to get the PUBLIC_POKEAPI environment variable.

// When import.meta.env.SSR === true
const data = await db(import.meta.env.DB_PASSWORD);

// When import.meta.env.SSR === false
const data = fetch(`${import.meta.env.PUBLIC_POKEAPI}/pokemon/squirtle`);

When using SSR, environment variables can be accessed at runtime based on the SSR adapter being used. With most adapters you can access environment variables with process.env, but some adapters work differently. For the Deno adapter, you will use Deno.env.get(). See how to access the Cloudflare runtime to handle environment variables when using the Cloudflare adapter. Astro will first check the server environment for variables, and if they don’t exist, Astro will look for them in .env files.

Type safe environment variables

The astro:env API lets you configure a type-safe schema for environment variables you have set. This allows you to indicate whether they should be available on the server or the client, and define their data type and additional properties.

Developing an adapter? See how to make an adapter compatible with astro:env.

Basic Usage

Define your schema

To configure a schema, add the env.schema option to your Astro config:

import { defineConfig } from 'astro/config'

export default defineConfig({
  env: {
    schema: {
      // ...
    }
  }
})

You can then register variables as a string, number, enum, or boolean using the envField helper. Define the kind of environment variable by providing a context (client or server) and access (secret or public) for each variable, and pass any additional properties such as optional or default in an object:

import { defineConfig, envField } from 'astro/config'

export default defineConfig({
  env: {
    schema: {
      API_URL: envField.string({ context: "client", access: "public", optional: true }),
      PORT: envField.number({ context: "server", access: "public", default: 4321 }),
      API_SECRET: envField.string({ context: "server", access: "secret" }),
    }
  }
})

Types will be generated for you when running astro dev or astro build, but you can run astro sync to generate types only.

Use variables from your schema

Import and use your defined variables from the appropriate /client or /server module:

---
import { API_URL } from "astro:env/client"
import { API_SECRET_TOKEN } from "astro:env/server"

const data = await fetch(`${API_URL}/users`, {
  method: "GET",
  headers: {
    "Content-Type": "application/json",
    "Authorization": `Bearer ${API_SECRET_TOKEN}`
  },
})
---

<script>
import { API_URL } from "astro:env/client"

fetch(`${API_URL}/ping`)
</script>

Variable types

There are three kinds of environment variables, determined by the combination of context (client or server) and access (secret or public) settings defined in your schema:

Public client variables: These variables end up in both your final client and server bundles, and can be accessed from both client and server through the astro:env/client module:
```
import { API_URL } from "astro:env/client"
```
Public server variables: These variables end up in your final server bundle and can be accessed on the server through the astro:env/server module:
```
import { PORT } from "astro:env/server"
```
Secret server variables: These variables are not part of your final bundle and can be accessed on the server through the astro:env/server module:
```
import { API_SECRET } from "astro:env/server"
```
By default, secrets are only validated at runtime. You can enable validating private variables on start by configuring validateSecrets: true.

Data types

There are currently four data types supported: strings, numbers, enums, and booleans:

import { envField } from "astro/config"

envField.string({
   // context & access
   optional: true,
   default: "foo",
})

envField.number({
   // context & access
   optional: true,
   default: 15,
})

envField.boolean({
   // context & access
   optional: true,
   default: true,
})

envField.enum({
   // context & access
   values: ['foo', 'bar', 'baz'],
   optional: true,
   default: 'baz',
})

For a complete list of validation fields, see the envField API reference.

Retrieving secrets dynamically

Despite defining your schema, you may want to retrieve the raw value of a given secret or to retrieve secrets not defined in your schema. In this case, you can use getSecret() exported from astro:env/server:

import {
   FOO, // boolean
   getSecret
} from 'astro:env/server'

getSecret('FOO') // string | undefined

Learn more in the API reference.

Limitations

astro:env is a virtual module which means it can only be used inside the Astro context. For example, you can use it in:
- Middlewares
- Astro routes and endpoints
- Astro components
- Framework components
- Modules
You cannot use it in the following and will have to resort to process.env:
- astro.config.mjs
- Scripts

@astrojs/cloudflare is a bit different than other adapters. Environment variables are scoped to the request, unlike Node.js where it’s global.

That means you always need to use secrets inside the request scope:

import { defineMiddleware } from "astro:middleware"
import { FOO, getSecret } from "astro:env"

console.log(FOO) // undefined
console.log(getSecret("FOO")) // undefined

export const onRequest = defineMiddleware((context, next) => {
   console.log(FOO) // boolean
   console.log(getSecret("FOO")) // string

   return next()
})

貢獻

社群

贊助者

贊助者

贊助者

贊助者

贊助者