Helpers/Accepts
Accepts Helper
Accepts Helper helps to handle Accept headers in the Requests.
Import
import { Hono } from 'hono'
import { accepts } from 'hono/accepts'`accepts()`
The accepts() function looks at the Accept header, such as Accept-Encoding and Accept-Language, and returns the proper value.
import { accepts } from 'hono/accepts'
app.get('/', (c) => {
const accept = accepts(c, {
header: 'Accept-Language',
supports: ['en', 'ja', 'zh'],
default: 'en',
})
return c.json({ lang: accept })
})`AcceptHeader` type
The definition of the AcceptHeader type is as follows.
export type AcceptHeader =
| 'Accept'
| 'Accept-Charset'
| 'Accept-Encoding'
| 'Accept-Language'
| 'Accept-Patch'
| 'Accept-Post'
| 'Accept-Ranges'Options
header: `AcceptHeader`
The target accept header.
supports: `string[]`
The header values which your application supports.
default: `string`
The default values.
match: `(accepts: Accept[], config: acceptsConfig) => string`
The custom match function.
Helpers/Adapter
Adapter Helper
The Adapter Helper provides a seamless way to interact with various platforms through a unified interface.
Import
import { Hono } from 'hono'
import { env, getRuntimeKey } from 'hono/adapter'`env()`
The env() function facilitates retrieving environment variables across different runtimes, extending beyond just Cloudflare Workers' Bindings. The value that can be retrieved with env(c) may be different for each runtimes.
import { env } from 'hono/adapter'
app.get('/env', (c) => {
// NAME is process.env.NAME on Node.js or Bun
// NAME is the value written in `wrangler.toml` on Cloudflare
const { NAME } = env<{ NAME: string }>(c)
return c.text(NAME)
})Supported Runtimes, Serverless Platforms and Cloud Services:
- Cloudflare Workers
wrangler.tomlwrangler.jsonc
- Deno
Deno.env.envfile
- Bun
Bun.envprocess.env
- Node.js
process.env
- Vercel
- AWS Lambda
- Lambda@Edge
Environment Variables on Lambda are not supported by Lambda@Edge, you need to use Lamdba@Edge event as an alternative. - Fastly Compute
On Fastly Compute, you can use the ConfigStore to manage user-defined data. - Netlify
On Netlify, you can use the Netlify Contexts to manage user-defined data.
Specify the runtime
You can specify the runtime to get environment variables by passing the runtime key as the second argument.
app.get('/env', (c) => {
const { NAME } = env<{ NAME: string }>(c, 'workerd')
return c.text(NAME)
})`getRuntimeKey()`
The getRuntimeKey() function returns the identifier of the current runtime.
app.get('/', (c) => {
if (getRuntimeKey() === 'workerd') {
return c.text('You are on Cloudflare')
} else if (getRuntimeKey() === 'bun') {
return c.text('You are on Bun')
}
...
})Available Runtimes Keys
Here are the available runtimes keys, unavailable runtime key runtimes may be supported and labeled as other, with some being inspired by WinterCG's Runtime Keys:
workerd- Cloudflare Workersdenobunnodeedge-light- Vercel Edge Functionsfastly- Fastly Computeother- Other unknown runtimes keys
Helpers/Conninfo
ConnInfo Helper
The ConnInfo Helper helps you to get the connection information. For example, you can get the client's remote address easily.
Import
::: code-group
import { Hono } from 'hono'
import { getConnInfo } from 'hono/cloudflare-workers'import { Hono } from 'hono'
import { getConnInfo } from 'hono/deno'import { Hono } from 'hono'
import { getConnInfo } from 'hono/bun'import { Hono } from 'hono'
import { getConnInfo } from 'hono/vercel'import { Hono } from 'hono'
import { getConnInfo } from 'hono/lambda-edge'import { Hono } from 'hono'
import { getConnInfo } from '@hono/node-server/conninfo':::
Usage
const app = new Hono()
app.get('/', (c) => {
const info = getConnInfo(c) // info is `ConnInfo`
return c.text(`Your remote address is ${info.remote.address}`)
})Type Definitions
The type definitions of the values that you can get from getConnInfo() are the following:
type AddressType = 'IPv6' | 'IPv4' | undefined
type NetAddrInfo = {
/**
* Transport protocol type
*/
transport?: 'tcp' | 'udp'
/**
* Transport port number
*/
port?: number
address?: string
addressType?: AddressType
} & (
| {
/**
* Host name such as IP Addr
*/
address: string
/**
* Host name type
*/
addressType: AddressType
}
| {}
)
/**
* HTTP Connection information
*/
interface ConnInfo {
/**
* Remote information
*/
remote: NetAddrInfo
}Helpers/Cookie
Cookie Helper
The Cookie Helper provides an easy interface to manage cookies, enabling developers to set, parse, and delete cookies seamlessly.
Import
import { Hono } from 'hono'
import {
deleteCookie,
getCookie,
getSignedCookie,
setCookie,
setSignedCookie,
generateCookie,
generateSignedCookie,
} from 'hono/cookie'Usage
Regular cookies
app.get('/cookie', (c) => {
setCookie(c, 'cookie_name', 'cookie_value')
const yummyCookie = getCookie(c, 'cookie_name')
deleteCookie(c, 'cookie_name')
const allCookies = getCookie(c)
// ...
})Signed cookies
NOTE: Setting and retrieving signed cookies returns a Promise due to the async nature of the WebCrypto API, which is used to create HMAC SHA-256 signatures.
app.get('/signed-cookie', (c) => {
const secret = 'secret' // make sure it's a large enough string to be secure
await setSignedCookie(c, 'cookie_name0', 'cookie_value', secret)
const fortuneCookie = await getSignedCookie(
c,
secret,
'cookie_name0'
)
deleteCookie(c, 'cookie_name0')
// `getSignedCookie` will return `false` for a specified cookie if the signature was tampered with or is invalid
const allSignedCookies = await getSignedCookie(c, secret)
// ...
})Cookie Generation
generateCookie and generateSignedCookie functions allow you to create cookie strings directly without setting them in the response headers.
`generateCookie`
// Basic cookie generation
const cookie = generateCookie('delicious_cookie', 'macha')
// Returns: 'delicious_cookie=macha; Path=/'
// Cookie with options
const cookie = generateCookie('delicious_cookie', 'macha', {
path: '/',
secure: true,
httpOnly: true,
domain: 'example.com',
})`generateSignedCookie`
// Basic signed cookie generation
const signedCookie = await generateSignedCookie(
'delicious_cookie',
'macha',
'secret chocolate chips'
)
// Signed cookie with options
const signedCookie = await generateSignedCookie(
'delicious_cookie',
'macha',
'secret chocolate chips',
{
path: '/',
secure: true,
httpOnly: true,
}
)Note: Unlike setCookie and setSignedCookie, these functions only generate the cookie strings. You need to manually set them in headers if needed.
Options
`setCookie` & `setSignedCookie`
- domain:
string - expires:
Date - httpOnly:
boolean - maxAge:
number - path:
string - secure:
boolean - sameSite:
'Strict'|'Lax'|'None' - priority:
'Low' | 'Medium' | 'High' - prefix:
secure|'host' - partitioned:
boolean
Example:
// Regular cookies
setCookie(c, 'great_cookie', 'banana', {
path: '/',
secure: true,
domain: 'example.com',
httpOnly: true,
maxAge: 1000,
expires: new Date(Date.UTC(2000, 11, 24, 10, 30, 59, 900)),
sameSite: 'Strict',
})
// Signed cookies
await setSignedCookie(
c,
'fortune_cookie',
'lots-of-money',
'secret ingredient',
{
path: '/',
secure: true,
domain: 'example.com',
httpOnly: true,
maxAge: 1000,
expires: new Date(Date.UTC(2000, 11, 24, 10, 30, 59, 900)),
sameSite: 'Strict',
}
)`deleteCookie`
- path:
string - secure:
boolean - domain:
string
Example:
deleteCookie(c, 'banana', {
path: '/',
secure: true,
domain: 'example.com',
})deleteCookie returns the deleted value:
const deletedCookie = deleteCookie(c, 'delicious_cookie')`__Secure-` and `__Host-` prefix
The Cookie helper supports __Secure- and __Host- prefix for cookies names.
If you want to verify if the cookie name has a prefix, specify the prefix option.
const securePrefixCookie = getCookie(c, 'yummy_cookie', 'secure')
const hostPrefixCookie = getCookie(c, 'yummy_cookie', 'host')
const securePrefixSignedCookie = await getSignedCookie(
c,
secret,
'fortune_cookie',
'secure'
)
const hostPrefixSignedCookie = await getSignedCookie(
c,
secret,
'fortune_cookie',
'host'
)Also, if you wish to specify a prefix when setting the cookie, specify a value for the prefix option.
setCookie(c, 'delicious_cookie', 'macha', {
prefix: 'secure', // or `host`
})
await setSignedCookie(
c,
'delicious_cookie',
'macha',
'secret choco chips',
{
prefix: 'secure', // or `host`
}
)Following the best practices
A New Cookie RFC (a.k.a cookie-bis) and CHIPS include some best practices for Cookie settings that developers should follow.
- RFC6265bis-13
Max-Age/Expireslimitation__Host-/__Secure-prefix limitation
- CHIPS-01
Partitionedlimitation
Hono is following the best practices.
The cookie helper will throw an Error when parsing cookies under the following conditions:
- The cookie name starts with
__Secure-, butsecureoption is not set. - The cookie name starts with
__Host-, butsecureoption is not set. - The cookie name starts with
__Host-, butpathis not/. - The cookie name starts with
__Host-, butdomainis set. - The
maxAgeoption value is greater than 400 days. - The
expiresoption value is 400 days later than the current time.
Helpers/Css
css Helper
The css helper - hono/css - is Hono's built-in CSS in JS(X).
You can write CSS in JSX in a JavaScript template literal named css. The return value of css will be the class name, which is set to the value of the class attribute. The <Style /> component will then contain the value of the CSS.
Import
import { Hono } from 'hono'
import { css, cx, keyframes, Style } from 'hono/css'`css`
You can write CSS in the css template literal. In this case, it uses headerClass as a value of the class attribute. Don't forget to add <Style /> as it contains the CSS content.
app.get('/', (c) => {
const headerClass = css`
background-color: orange;
color: white;
padding: 1rem;
`
return c.html(
<html>
<head>
<Style />
</head>
<h1 class={headerClass}>Hello!</h1>
</html>
)
})You can style pseudo-classes like :hover by using the nesting selector, &:
const buttonClass = css`
background-color: #fff;
&:hover {
background-color: red;
}
`Extending
You can extend the CSS definition by embedding the class name.
const baseClass = css`
color: white;
background-color: blue;
`
const header1Class = css`
${baseClass}
font-size: 3rem;
`
const header2Class = css`
${baseClass}
font-size: 2rem;
`In addition, the syntax of ${baseClass} {} enables nesting classes.
const headerClass = css`
color: white;
background-color: blue;
`
const containerClass = css`
${headerClass} {
h1 {
font-size: 3rem;
}
}
`
return c.render(
<header class={headerClass}>
<h1>Hello!</h1>
</header>
)Global styles
A pseudo-selector called :-hono-global allows you to define global styles.
const globalClass = css`
:-hono-global {
html {
font-family: Arial, Helvetica, sans-serif;
}
}
`
return c.render(
<h1>Hello!</h1>
Today is a good day.
)Or you can write CSS in the <Style /> component with the css literal.
export const renderer = jsxRenderer(({ children, title }) => {
return (
<html>
<head>
<Style>{css`
html {
font-family: Arial, Helvetica, sans-serif;
}
`}</Style>
<title>{title}</title>
</head>
{children}
</html>
)
})`keyframes`
You can use keyframes to write the contents of @keyframes. In this case, fadeInAnimation will be the name of the animation
const fadeInAnimation = keyframes`
from {
opacity: 0;
}
to {
opacity: 1;
}
`
const headerClass = css`
animation-name: ${fadeInAnimation};
animation-duration: 2s;
`
const Header = () => <a class={headerClass}>Hello!</a>`cx`
The cx composites the two class names.
const buttonClass = css`
border-radius: 10px;
`
const primaryClass = css`
background: orange;
`
const Button = () => (
<a class={cx(buttonClass, primaryClass)}>Click!</a>
)It can also compose simple strings.
const Header = () => <a class={cx('h1', primaryClass)}>Hi</a>Usage in combination with [Secure Headers](/docs/middleware/builtin/secure-headers) middleware
If you want to use the css helpers in combination with the Secure Headers middleware, you can add the nonce attribute to the <Style nonce={c.get('secureHeadersNonce')} /> to avoid Content-Security-Policy caused by the css helpers.
import { secureHeaders, NONCE } from 'hono/secure-headers'
app.get(
'*',
secureHeaders({
contentSecurityPolicy: {
// Set the pre-defined nonce value to `styleSrc`:
styleSrc: [NONCE],
},
})
)
app.get('/', (c) => {
const headerClass = css`
background-color: orange;
color: white;
padding: 1rem;
`
return c.html(
<html>
<head>
{/* Set the `nonce` attribute on the css helpers `style` and `script` elements */}
<Style nonce={c.get('secureHeadersNonce')} />
</head>
<h1 class={headerClass}>Hello!</h1>
</html>
)
})Tips
If you use VS Code, you can use vscode-styled-components for Syntax highlighting and IntelliSense for css tagged literals.
Helpers/Dev
Dev Helper
Dev Helper provides useful methods you can use in development.
import { Hono } from 'hono'
import { getRouterName, showRoutes } from 'hono/dev'`getRouterName()`
You can get the name of the currently used router with getRouterName().
const app = new Hono()
// ...
console.log(getRouterName(app))`showRoutes()`
showRoutes() function displays the registered routes in your console.
Consider an application like the following:
const app = new Hono().basePath('/v1')
app.get('/posts', (c) => {
// ...
})
app.get('/posts/:id', (c) => {
// ...
})
app.post('/posts', (c) => {
// ...
})
showRoutes(app, {
verbose: true,
})When this application starts running, the routes will be shown in your console as follows:
GET /v1/posts
GET /v1/posts/:id
POST /v1/postsOptions
verbose: `boolean`
When set to true, it displays verbose information.
colorize: `boolean`
When set to false, the output will not be colored.
Helpers/Factory
Factory Helper
The Factory Helper provides useful functions for creating Hono's components such as Middleware. Sometimes it's difficult to set the proper TypeScript types, but this helper facilitates that.
Import
import { Hono } from 'hono'
import { createFactory, createMiddleware } from 'hono/factory'`createFactory()`
createFactory() will create an instance of the Factory class.
import { createFactory } from 'hono/factory'
const factory = createFactory()You can pass your Env types as Generics:
type Env = {
Variables: {
foo: string
}
}
const factory = createFactory<Env>()Options
defaultAppOptions: `HonoOptions`
The default options to pass to the Hono application created by createApp().
const factory = createFactory({
defaultAppOptions: { strict: false },
})
const app = factory.createApp() // `strict: false` is applied`createMiddleware()`
createMiddleware() is shortcut of factory.createMiddleware().
This function will create your custom middleware.
const messageMiddleware = createMiddleware(async (c, next) => {
await next()
c.res.headers.set('X-Message', 'Good morning!')
})Tip: If you want to get an argument like message, you can create it as a function like the following.
const messageMiddleware = (message: string) => {
return createMiddleware(async (c, next) => {
await next()
c.res.headers.set('X-Message', message)
})
}
app.use(messageMiddleware('Good evening!'))`factory.createHandlers()`
createHandlers() helps to define handlers in a different place than app.get('/').
import { createFactory } from 'hono/factory'
import { logger } from 'hono/logger'
// ...
const factory = createFactory()
const middleware = factory.createMiddleware(async (c, next) => {
c.set('foo', 'bar')
await next()
})
const handlers = factory.createHandlers(logger(), middleware, (c) => {
return c.json(c.var.foo)
})
app.get('/api', ...handlers)`factory.createApp()`
createApp() helps to create an instance of Hono with the proper types. If you use this method with createFactory(), you can avoid redundancy in the definition of the Env type.
If your application is like this, you have to set the Env in two places:
import { createMiddleware } from 'hono/factory'
type Env = {
Variables: {
myVar: string
}
}
// 1. Set the `Env` to `new Hono()`
const app = new Hono<Env>()
// 2. Set the `Env` to `createMiddleware()`
const mw = createMiddleware<Env>(async (c, next) => {
await next()
})
app.use(mw)By using createFactory() and createApp(), you can set the Env only in one place.
import { createFactory } from 'hono/factory'
// ...
// Set the `Env` to `createFactory()`
const factory = createFactory<Env>()
const app = factory.createApp()
// factory also has `createMiddleware()`
const mw = factory.createMiddleware(async (c, next) => {
await next()
})createFactory() can receive the initApp option to initialize an app created by createApp(). The following is an example that uses the option.
// factory-with-db.ts
type Env = {
Bindings: {
MY_DB: D1Database
}
Variables: {
db: DrizzleD1Database
}
}
export default createFactory<Env>({
initApp: (app) => {
app.use(async (c, next) => {
const db = drizzle(c.env.MY_DB)
c.set('db', db)
await next()
})
},
})// crud.ts
import factoryWithDB from './factory-with-db'
const app = factoryWithDB.createApp()
app.post('/posts', (c) => {
c.var.db.insert()
// ...
})Helpers/Html
html Helper
The html Helper lets you write HTML in JavaScript template literal with a tag named html. Using raw(), the content will be rendered as is. You have to escape these strings by yourself.
Import
import { Hono } from 'hono'
import { html, raw } from 'hono/html'`html`
const app = new Hono()
app.get('/:username', (c) => {
const { username } = c.req.param()
return c.html(
html`<!doctype html>
<h1>Hello! ${username}!</h1>`
)
})Insert snippets into JSX
Insert the inline script into JSX:
app.get('/', (c) => {
return c.html(
<html>
<head>
<title>Test Site</title>
{html`
<script>
// No need to use dangerouslySetInnerHTML.
// If you write it here, it will not be escaped.
</script>
`}
</head>
Hello!
</html>
)
})Act as functional component
Since html returns an HtmlEscapedString, it can act as a fully functional component without using JSX.
Use `html` to speed up the process instead of `memo`
const Footer = () => html`
<footer>
<address>My Address...</address>
</footer>
`Receives props and embeds values
interface SiteData {
title: string
image: string
children?: any
}
const Layout = (props: SiteData) => html`
<html>
<head>
<meta charset="UTF-8">
<title>${props.title}</title>
<meta name="description" content="${props.description}">
<head prefix="og: http://ogp.me/ns#">
<meta property="og:type" content="article">
<meta property="og:title" content="${props.title}">
<meta property="og:image" content="${props.image}">
</head>
${props.children}
</html>
`
const Content = (props: { siteData: SiteData; name: string }) => (
<Layout {...props.siteData}>
<h1>Hello {props.name}</h1>
</Layout>
)
app.get('/', (c) => {
const props = {
name: 'World',
siteData: {
title: 'Hello <> World',
image: 'https://example.com/image.png',
},
}
return c.html(<Content {...props} />)
})`raw()`
app.get('/', (c) => {
const name = 'John "Johnny" Smith'
return c.html(html`I'm ${raw(name)}.`)
})Tips
Thanks to these libraries, Visual Studio Code and vim also interprets template literals as HTML, allowing syntax highlighting and formatting to be applied.
- https://marketplace.visualstudio.com/items?itemName=bierner.lit-html
- https://github.com/MaxMEllon/vim-jsx-pretty
Helpers/Jwt
JWT Authentication Helper
This helper provides functions for encoding, decoding, signing, and verifying JSON Web Tokens (JWTs). JWTs are commonly used for authentication and authorization purposes in web applications. This helper offers robust JWT functionality with support for various cryptographic algorithms.
Import
To use this helper, you can import it as follows:
import { decode, sign, verify } from 'hono/jwt'::: info
JWT Middleware also import the jwt function from the hono/jwt.
:::
`sign()`
This function generates a JWT token by encoding a payload and signing it using the specified algorithm and secret.
sign(
payload: unknown,
secret: string,
alg?: 'HS256';
): Promise<string>;Example
import { sign } from 'hono/jwt'
const payload = {
sub: 'user123',
role: 'admin',
exp: Math.floor(Date.now() / 1000) + 60 * 5, // Token expires in 5 minutes
}
const secret = 'mySecretKey'
const token = await sign(payload, secret)Options
payload: `unknown`
The JWT payload to be signed. You can include other claims like in Payload Validation.
secret: `string`
The secret key used for JWT verification or signing.
alg: [AlgorithmTypes](#supported-algorithmtypes)
The algorithm used for JWT signing or verification. The default is HS256.
`verify()`
This function checks if a JWT token is genuine and still valid. It ensures the token hasn't been altered and checks validity only if you added Payload Validation.
verify(
token: string,
secret: string,
alg?: 'HS256';
issuer?: string | RegExp;
): Promise<any>;Example
import { verify } from 'hono/jwt'
const tokenToVerify = 'token'
const secretKey = 'mySecretKey'
const decodedPayload = await verify(tokenToVerify, secretKey)
console.log(decodedPayload)Options
token: `string`
The JWT token to be verified.
secret: `string`
The secret key used for JWT verification or signing.
alg: [AlgorithmTypes](#supported-algorithmtypes)
The algorithm used for JWT signing or verification. The default is HS256.
issuer: `string | RegExp`
The expected issuer used for JWT verification.
`decode()`
This function decodes a JWT token without performing signature verification. It extracts and returns the header and payload from the token.
decode(token: string): { header: any; payload: any };Example
import { decode } from 'hono/jwt'
// Decode the JWT token
const tokenToDecode =
'eyJhbGciOiAiSFMyNTYiLCAidHlwIjogIkpXVCJ9.eyJzdWIiOiAidXNlcjEyMyIsICJyb2xlIjogImFkbWluIn0.JxUwx6Ua1B0D1B0FtCrj72ok5cm1Pkmr_hL82sd7ELA'
const { header, payload } = decode(tokenToDecode)
console.log('Decoded Header:', header)
console.log('Decoded Payload:', payload)Options
token: `string`
The JWT token to be decoded.
The
decodefunction allows you to inspect the header and payload of a JWT token without performing verification. This can be useful for debugging or extracting information from JWT tokens.
Payload Validation
When verifying a JWT token, the following payload validations are performed:
exp: The token is checked to ensure it has not expired.nbf: The token is checked to ensure it is not being used before a specified time.iat: The token is checked to ensure it is not issued in the future.iss: The token is checked to ensure it has been issued by a trusted issuer.
Please ensure that your JWT payload includes these fields, as an object, if you intend to perform these checks during verification.
Custom Error Types
The module also defines custom error types to handle JWT-related errors.
JwtAlgorithmNotImplemented: Indicates that the requested JWT algorithm is not implemented.JwtTokenInvalid: Indicates that the JWT token is invalid.JwtTokenNotBefore: Indicates that the token is being used before its valid date.JwtTokenExpired: Indicates that the token has expired.JwtTokenIssuedAt: Indicates that the "iat" claim in the token is incorrect.JwtTokenIssuer: Indicates that the "iss" claim in the token is incorrect.JwtTokenSignatureMismatched: Indicates a signature mismatch in the token.
Supported AlgorithmTypes
The module supports the following JWT cryptographic algorithms:
HS256: HMAC using SHA-256HS384: HMAC using SHA-384HS512: HMAC using SHA-512RS256: RSASSA-PKCS1-v1_5 using SHA-256RS384: RSASSA-PKCS1-v1_5 using SHA-384RS512: RSASSA-PKCS1-v1_5 using SHA-512PS256: RSASSA-PSS using SHA-256 and MGF1 with SHA-256PS384: RSASSA-PSS using SHA-386 and MGF1 with SHA-386PS512: RSASSA-PSS using SHA-512 and MGF1 with SHA-512ES256: ECDSA using P-256 and SHA-256ES384: ECDSA using P-384 and SHA-384ES512: ECDSA using P-521 and SHA-512EdDSA: EdDSA using Ed25519
Helpers/Proxy
Proxy Helper
Proxy Helper provides useful functions when using Hono application as a (reverse) proxy.
Import
import { Hono } from 'hono'
import { proxy } from 'hono/proxy'`proxy()`
proxy() is a fetch() API wrapper for proxy. The parameters and return value are the same as for fetch() (except for the proxy-specific options).
The Accept-Encoding header is replaced with an encoding that the current runtime can handle. Unnecessary response headers are deleted, and a Response object is returned that you can return as a response from the handler.
Examples
Simple usage:
app.get('/proxy/:path', (c) => {
return proxy(`http://${originServer}/${c.req.param('path')}`)
})Complicated usage:
app.get('/proxy/:path', async (c) => {
const res = await proxy(
`http://${originServer}/${c.req.param('path')}`,
{
headers: {
...c.req.header(), // optional, specify only when forwarding all the request data (including credentials) is necessary.
'X-Forwarded-For': '127.0.0.1',
'X-Forwarded-Host': c.req.header('host'),
Authorization: undefined, // do not propagate request headers contained in c.req.header('Authorization')
},
}
)
res.headers.delete('Set-Cookie')
return res
})Or you can pass the c.req as a parameter.
app.all('/proxy/:path', (c) => {
return proxy(`http://${originServer}/${c.req.param('path')}`, {
...c.req, // optional, specify only when forwarding all the request data (including credentials) is necessary.
headers: {
...c.req.header(),
'X-Forwarded-For': '127.0.0.1',
'X-Forwarded-Host': c.req.header('host'),
Authorization: undefined, // do not propagate request headers contained in c.req.header('Authorization')
},
})
})You can override the default global fetch function with the customFetch option:
app.get('/proxy', (c) => {
return proxy('https://example.com/', {
customFetch,
})
})Connection Header Processing
By default, proxy() ignores the Connection header to prevent Hop-by-Hop Header Injection attacks. You can enable strict RFC 9110 compliance with the strictConnectionProcessing option:
// Default behavior (recommended for untrusted clients)
app.get('/proxy/:path', (c) => {
return proxy(`http://${originServer}/${c.req.param('path')}`, c.req)
})
// Strict RFC 9110 compliance (use only in trusted environments)
app.get('/internal-proxy/:path', (c) => {
return proxy(`http://${internalServer}/${c.req.param('path')}`, {
...c.req,
strictConnectionProcessing: true,
})
})`ProxyFetch`
The type of proxy() is defined as ProxyFetch and is as follows
interface ProxyRequestInit extends Omit<RequestInit, 'headers'> {
raw?: Request
customFetch?: (request: Request) => Promise<Response>
strictConnectionProcessing?: boolean
headers?:
| HeadersInit
| [string, string][]
| Record<RequestHeader, string | undefined>
| Record<string, string | undefined>
}
interface ProxyFetch {
(
input: string | URL | Request,
init?: ProxyRequestInit
): Promise<Response>
}Helpers/Route
Route Helper
The Route Helper provides enhanced routing information for debugging and middleware development. It allows you to access detailed information about matched routes and the current route being processed.
Import
import { Hono } from 'hono'
import {
matchedRoutes,
routePath,
baseRoutePath,
basePath,
} from 'hono/route'Usage
Basic route information
const app = new Hono()
app.get('/posts/:id', (c) => {
const currentPath = routePath(c) // '/posts/:id'
const routes = matchedRoutes(c) // Array of matched routes
return c.json({
path: currentPath,
totalRoutes: routes.length,
})
})Working with sub-applications
const app = new Hono()
const apiApp = new Hono()
apiApp.get('/posts/:id', (c) => {
return c.json({
routePath: routePath(c), // '/posts/:id'
baseRoutePath: baseRoutePath(c), // '/api'
basePath: basePath(c), // '/api' (with actual params)
})
})
app.route('/api', apiApp)`matchedRoutes()`
Returns an array of all routes that matched the current request, including middleware.
app.all('/api/*', (c, next) => {
console.log('API middleware')
return next()
})
app.get('/api/users/:id', (c) => {
const routes = matchedRoutes(c)
// Returns: [
// { method: 'ALL', path: '/api/*', handler: [Function] },
// { method: 'GET', path: '/api/users/:id', handler: [Function] }
// ]
return c.json({ routes: routes.length })
})`routePath()`
Returns the route path pattern registered for the current handler.
app.get('/posts/:id', (c) => {
console.log(routePath(c)) // '/posts/:id'
return c.text('Post details')
})Using with index parameter
You can optionally pass an index parameter to get the route path at a specific position, similar to Array.prototype.at().
app.all('/api/*', (c, next) => {
return next()
})
app.get('/api/users/:id', (c) => {
console.log(routePath(c, 0)) // '/api/*' (first matched route)
console.log(routePath(c, -1)) // '/api/users/:id' (last matched route)
return c.text('User details')
})`baseRoutePath()`
Returns the base path pattern of the current route as specified in routing.
const subApp = new Hono()
subApp.get('/posts/:id', (c) => {
return c.text(baseRoutePath(c)) // '/:sub'
})
app.route('/:sub', subApp)Using with index parameter
You can optionally pass an index parameter to get the base route path at a specific
position, similar to Array.prototype.at().
app.all('/api/*', (c, next) => {
return next()
})
const subApp = new Hono()
subApp.get('/users/:id', (c) => {
console.log(baseRoutePath(c, 0)) // '/' (first matched route)
console.log(baseRoutePath(c, -1)) // '/api' (last matched route)
return c.text('User details')
})
app.route('/api', subApp)`basePath()`
Returns the base path with embedded parameters from the actual request.
const subApp = new Hono()
subApp.get('/posts/:id', (c) => {
return c.text(basePath(c)) // '/api' (for request to '/api/posts/123')
})
app.route('/:sub', subApp)Helpers/Ssg
SSG Helper
SSG Helper generates a static site from your Hono application. It will retrieve the contents of registered routes and save them as static files.
Usage
Manual
If you have a simple Hono application like the following:
// index.tsx
const app = new Hono()
app.get('/', (c) => c.html('Hello, World!'))
app.use('/about', async (c, next) => {
c.setRenderer((content) => {
return c.html(
<html>
<head />
{content}
</html>
)
})
await next()
})
app.get('/about', (c) => {
return c.render(
<>
<title>Hono SSG Page</title>Hello!
</>
)
})
export default appFor Node.js, create a build script like this:
// build.ts
import app from './index'
import { toSSG } from 'hono/ssg'
import fs from 'fs/promises'
toSSG(app, fs)By executing the script, the files will be output as follows:
ls ./static
about.html index.htmlVite Plugin
Using the @hono/vite-ssg Vite Plugin, you can easily handle the process.
For more details, see here:
https://github.com/honojs/vite-plugins/tree/main/packages/ssg
toSSG
toSSG is the main function for generating static sites, taking an application and a filesystem module as arguments. It is based on the following:
Input
The arguments for toSSG are specified in ToSSGInterface.
export interface ToSSGInterface {
(
app: Hono,
fsModule: FileSystemModule,
options?: ToSSGOptions
): Promise<ToSSGResult>
}appspecifiesnew Hono()with registered routes.fsspecifies the following object, assumingnode:fs/promise.
export interface FileSystemModule {
writeFile(path: string, data: string | Uint8Array): Promise<void>
mkdir(
path: string,
options: { recursive: boolean }
): Promise<void | string>
}Using adapters for Deno and Bun
If you want to use SSG on Deno or Bun, a toSSG function is provided for each file system.
For Deno:
import { toSSG } from 'hono/deno'
toSSG(app) // The second argument is an option typed `ToSSGOptions`.For Bun:
import { toSSG } from 'hono/bun'
toSSG(app) // The second argument is an option typed `ToSSGOptions`.Options
Options are specified in the ToSSGOptions interface.
export interface ToSSGOptions {
dir?: string
concurrency?: number
extensionMap?: Record<string, string>
plugins?: SSGPlugin[]
}diris the output destination for Static files. The default value is./static.concurrencyis the concurrent number of files to be generated at the same time. The default value is2.extensionMapis a map containing theContent-Typeas a key and the string of the extension as a value. This is used to determine the file extension of the output file.pluginsis an array of SSG plugins that extend the functionality of the static site generation process.
Output
toSSG returns the result in the following Result type.
export interface ToSSGResult {
success: boolean
files: string[]
error?: Error
}Generate File
Route and Filename
The following rules apply to the registered route information and the generated file name. The default ./static behaves as follows:
/->./static/index.html/path->./static/path.html/path/->./static/path/index.html
File Extension
The file extension depends on the Content-Type returned by each route. For example, responses from c.html are saved as .html.
If you want to customize the file extensions, set the extensionMap option.
import { toSSG, defaultExtensionMap } from 'hono/ssg'
// Save `application/x-html` content with `.html`
toSSG(app, fs, {
extensionMap: {
'application/x-html': 'html',
...defaultExtensionMap,
},
})Note that paths ending with a slash are saved as index.ext regardless of the extension.
// save to ./static/html/index.html
app.get('/html/', (c) => c.html('html'))
// save to ./static/text/index.txt
app.get('/text/', (c) => c.text('text'))Middleware
Introducing built-in middleware that supports SSG.
ssgParams
You can use an API like generateStaticParams of Next.js.
Example:
app.get(
'/shops/:id',
ssgParams(async () => {
const shops = await getShops()
return shops.map((shop) => ({ id: shop.id }))
}),
async (c) => {
const shop = await getShop(c.req.param('id'))
if (!shop) {
return c.notFound()
}
return c.render(
<h1>{shop.name}</h1>
)
}
)disableSSG
Routes with the disableSSG middleware set are excluded from static file generation by toSSG.
app.get('/api', disableSSG(), (c) => c.text('an-api'))onlySSG
Routes with the onlySSG middleware set will be overridden by c.notFound() after toSSG execution.
app.get('/static-page', onlySSG(), (c) => c.html(<h1>Welcome to my site</h1>))Plugins
Plugins allow you to extend the functionality of the static site generation process. They use hooks to customize the generation process at different stages.
Default Plugin
By default, toSSG uses defaultPlugin which skips non-200 status responses (like redirects, errors, or 404s). This prevents generating files for unsuccessful responses.
import { toSSG, defaultPlugin } from 'hono/ssg'
// defaultPlugin is automatically applied when no plugins specified
toSSG(app, fs)
// Equivalent to:
toSSG(app, fs, { plugins: [defaultPlugin] })If you specify custom plugins, defaultPlugin is not automatically included. To keep the default behavior while adding custom plugins, explicitly include it:
toSSG(app, fs, {
plugins: [defaultPlugin, myCustomPlugin],
})Hook Types
Plugins can use the following hooks to customize the toSSG process:
export type BeforeRequestHook = (req: Request) => Request | false
export type AfterResponseHook = (res: Response) => Response | false
export type AfterGenerateHook = (
result: ToSSGResult
) => void | Promise<void>- BeforeRequestHook: Called before processing each request. Return
falseto skip the route. - AfterResponseHook: Called after receiving each response. Return
falseto skip file generation. - AfterGenerateHook: Called after the entire generation process completes.
Plugin Interface
export interface SSGPlugin {
beforeRequestHook?: BeforeRequestHook | BeforeRequestHook[]
afterResponseHook?: AfterResponseHook | AfterResponseHook[]
afterGenerateHook?: AfterGenerateHook | AfterGenerateHook[]
}Basic Plugin Examples
Filter only GET requests:
const getOnlyPlugin: SSGPlugin = {
beforeRequestHook: (req) => {
if (req.method === 'GET') {
return req
}
return false
},
}Filter by status code:
const statusFilterPlugin: SSGPlugin = {
afterResponseHook: (res) => {
if (res.status === 200 || res.status === 500) {
return res
}
return false
},
}Log generated files:
const logFilesPlugin: SSGPlugin = {
afterGenerateHook: (result) => {
if (result.files) {
result.files.forEach((file) => console.log(file))
}
},
}Advanced Plugin Example
Here's an example of creating a sitemap plugin that generates a sitemap.xml file:
// plugins.ts
import fs from 'node:fs/promises'
import path from 'node:path'
import type { SSGPlugin } from 'hono/ssg'
import { DEFAULT_OUTPUT_DIR } from 'hono/ssg'
export const sitemapPlugin = (baseURL: string): SSGPlugin => {
return {
afterGenerateHook: (result, fsModule, options) => {
const outputDir = options?.dir ?? DEFAULT_OUTPUT_DIR
const filePath = path.join(outputDir, 'sitemap.xml')
const urls = result.files.map((file) =>
new URL(file, baseURL).toString()
)
const siteMapText = `<?xml version="1.0" encoding="UTF-8"?>
${urls.map((url) => `<loc>${url}</loc>`).join('\n')}
`
fsModule.writeFile(filePath, siteMapText)
},
}
}Applying plugins:
import app from './index'
import { toSSG } from 'hono/ssg'
import { sitemapPlugin } from './plugins'
toSSG(app, fs, {
plugins: [
getOnlyPlugin,
statusFilterPlugin,
logFilesPlugin,
sitemapPlugin('https://example.com'),
],
})Helpers/Streaming
Streaming Helper
The Streaming Helper provides methods for streaming responses.
Import
import { Hono } from 'hono'
import { stream, streamText, streamSSE } from 'hono/streaming'`stream()`
It returns a simple streaming response as Response object.
app.get('/stream', (c) => {
return stream(c, async (stream) => {
// Write a process to be executed when aborted.
stream.onAbort(() => {
console.log('Aborted!')
})
// Write a Uint8Array.
await stream.write(new Uint8Array([0x48, 0x65, 0x6c, 0x6c, 0x6f]))
// Pipe a readable stream.
await stream.pipe(anotherReadableStream)
})
})`streamText()`
It returns a streaming response with Content-Type:text/plain, Transfer-Encoding:chunked, and X-Content-Type-Options:nosniff headers.
app.get('/streamText', (c) => {
return streamText(c, async (stream) => {
// Write a text with a new line ('\n').
await stream.writeln('Hello')
// Wait 1 second.
await stream.sleep(1000)
// Write a text without a new line.
await stream.write(`Hono!`)
})
})::: warning
If you are developing an application for Cloudflare Workers, a streaming may not work well on Wrangler. If so, add Identity for Content-Encoding header.
app.get('/streamText', (c) => {
c.header('Content-Encoding', 'Identity')
return streamText(c, async (stream) => {
// ...
})
}):::
`streamSSE()`
It allows you to stream Server-Sent Events (SSE) seamlessly.
const app = new Hono()
let id = 0
app.get('/sse', async (c) => {
return streamSSE(c, async (stream) => {
while (true) {
const message = `It is ${new Date().toISOString()}`
await stream.writeSSE({
data: message,
event: 'time-update',
id: String(id++),
})
await stream.sleep(1000)
}
})
})Error Handling
The third argument of the streaming helper is an error handler.
This argument is optional, if you don't specify it, the error will be output as a console error.
app.get('/stream', (c) => {
return stream(
c,
async (stream) => {
// Write a process to be executed when aborted.
stream.onAbort(() => {
console.log('Aborted!')
})
// Write a Uint8Array.
await stream.write(
new Uint8Array([0x48, 0x65, 0x6c, 0x6c, 0x6f])
)
// Pipe a readable stream.
await stream.pipe(anotherReadableStream)
},
(err, stream) => {
stream.writeln('An error occurred!')
console.error(err)
}
)
})The stream will be automatically closed after the callbacks are executed.
::: warning
If the callback function of the streaming helper throws an error, the onError event of Hono will not be triggered.
onError is a hook to handle errors before the response is sent and overwrite the response. However, when the callback function is executed, the stream has already started, so it cannot be overwritten.
:::
Helpers/Testing
Testing Helper
The Testing Helper provides functions to make testing of Hono applications easier.
Import
import { Hono } from 'hono'
import { testClient } from 'hono/testing'`testClient()`
The testClient() function takes an instance of Hono as its first argument and returns an object typed according to your Hono application's routes, similar to the Hono Client. This allows you to call your defined routes in a type-safe manner with editor autocompletion within your tests.
Important Note on Type Inference:
For the testClient to correctly infer the types of your routes and provide autocompletion, you must define your routes using chained methods directly on the Hono instance.
The type inference relies on the type flowing through the chained .get(), .post(), etc., calls. If you define routes separately after creating the Hono instance (like the common pattern shown in the "Hello World" example: const app = new Hono(); app.get(...)), the testClient will not have the necessary type information for specific routes, and you won't get the type-safe client features.
Example:
This example works because the .get() method is chained directly onto the new Hono() call:
// index.ts
const app = new Hono().get('/search', (c) => {
const query = c.req.query('q')
return c.json({ query: query, results: ['result1', 'result2'] })
})
export default app// index.test.ts
import { Hono } from 'hono'
import { testClient } from 'hono/testing'
import { describe, it, expect } from 'vitest' // Or your preferred test runner
import app from './app'
describe('Search Endpoint', () => {
// Create the test client from the app instance
const client = testClient(app)
it('should return search results', async () => {
// Call the endpoint using the typed client
// Notice the type safety for query parameters (if defined in the route)
// and the direct access via .$get()
const res = await client.search.$get({
query: { q: 'hono' },
})
// Assertions
expect(res.status).toBe(200)
expect(await res.json()).toEqual({
query: 'hono',
results: ['result1', 'result2'],
})
})
})To include headers in your test, pass them as the second parameter in the call. The second parameter can also take an init property as a RequestInit object, allowing you to set headers, method, body, etc. Learn more about the init property here.
// index.test.ts
import { Hono } from 'hono'
import { testClient } from 'hono/testing'
import { describe, it, expect } from 'vitest' // Or your preferred test runner
import app from './app'
describe('Search Endpoint', () => {
// Create the test client from the app instance
const client = testClient(app)
it('should return search results', async () => {
// Include the token in the headers and set the content type
const token = 'this-is-a-very-clean-token'
const res = await client.search.$get(
{
query: { q: 'hono' },
},
{
headers: {
Authorization: `Bearer ${token}`,
'Content-Type': `application/json`,
},
}
)
// Assertions
expect(res.status).toBe(200)
expect(await res.json()).toEqual({
query: 'hono',
results: ['result1', 'result2'],
})
})
})Helpers/Websocket
WebSocket Helper
WebSocket Helper is a helper for server-side WebSockets in Hono applications.
Currently Cloudflare Workers / Pages, Deno, and Bun adapters are available.
Import
::: code-group
import { Hono } from 'hono'
import { upgradeWebSocket } from 'hono/cloudflare-workers'import { Hono } from 'hono'
import { upgradeWebSocket } from 'hono/deno'import { Hono } from 'hono'
import { upgradeWebSocket, websocket } from 'hono/bun'
// ...
export default {
fetch: app.fetch,
websocket,
}:::
If you use Node.js, you can use @hono/node-ws.
`upgradeWebSocket()`
upgradeWebSocket() returns a handler for handling WebSocket.
const app = new Hono()
app.get(
'/ws',
upgradeWebSocket((c) => {
return {
onMessage(event, ws) {
console.log(`Message from client: ${event.data}`)
ws.send('Hello from server!')
},
onClose: () => {
console.log('Connection closed')
},
}
})
)Available events:
onOpen- Currently, Cloudflare Workers does not support it.onMessageonCloseonError
::: warning
If you use middleware that modifies headers (e.g., applying CORS) on a route that uses WebSocket Helper, you may encounter an error saying you can't modify immutable headers. This is because upgradeWebSocket() also changes headers internally.
Therefore, please be cautious if you are using WebSocket Helper and middleware at the same time.
:::
RPC-mode
Handlers defined with WebSocket Helper support RPC mode.
// server.ts
const wsApp = app.get(
'/ws',
upgradeWebSocket((c) => {
//...
})
)
export type WebSocketApp = typeof wsApp
// client.ts
const client = hc<WebSocketApp>('http://localhost:8787')
const socket = client.ws.$ws() // A WebSocket object for a clientExamples
See the examples using WebSocket Helper.
Server and Client
// server.ts
import { Hono } from 'hono'
import { upgradeWebSocket } from 'hono/cloudflare-workers'
const app = new Hono().get(
'/ws',
upgradeWebSocket(() => {
return {
onMessage: (event) => {
console.log(event.data)
},
}
})
)
export default app// client.ts
import { hc } from 'hono/client'
import type app from './server'
const client = hc<typeof app>('http://localhost:8787')
const ws = client.ws.$ws(0)
ws.addEventListener('open', () => {
setInterval(() => {
ws.send(new Date().toString())
}, 1000)
})Bun with JSX
import { Hono } from 'hono'
import { upgradeWebSocket, websocket } from 'hono/bun'
import { html } from 'hono/html'
const app = new Hono()
app.get('/', (c) => {
return c.html(
<html>
<head>
<meta charset='UTF-8' />
</head>
{html`
<script>
const ws = new WebSocket('ws://localhost:3000/ws')
const $nowTime = document.getElementById('now-time')
ws.onmessage = (event) => {
$nowTime.textContent = event.data
}
</script>
`}
</html>
)
})
const ws = app.get(
'/ws',
upgradeWebSocket((c) => {
let intervalId
return {
onOpen(_event, ws) {
intervalId = setInterval(() => {
ws.send(new Date().toString())
}, 200)
},
onClose() {
clearInterval(intervalId)
},
}
})
)
export default {
fetch: app.fetch,
websocket,
}Guides/Best Practices
Best Practices
Hono is very flexible. You can write your app as you like.
However, there are best practices that are better to follow.
Don't make "Controllers" when possible
When possible, you should not create "Ruby on Rails-like Controllers".
// 🙁
// A RoR-like Controller
const booksList = (c: Context) => {
return c.json('list books')
}
app.get('/books', booksList)The issue is related to types. For example, the path parameter cannot be inferred in the Controller without writing complex generics.
// 🙁
// A RoR-like Controller
const bookPermalink = (c: Context) => {
const id = c.req.param('id') // Can't infer the path param
return c.json(`get ${id}`)
}Therefore, you don't need to create RoR-like controllers and should write handlers directly after path definitions.
// 😃
app.get('/books/:id', (c) => {
const id = c.req.param('id') // Can infer the path param
return c.json(`get ${id}`)
})`factory.createHandlers()` in `hono/factory`
If you still want to create a RoR-like Controller, use factory.createHandlers() in hono/factory. If you use this, type inference will work correctly.
import { createFactory } from 'hono/factory'
import { logger } from 'hono/logger'
// ...
// 😃
const factory = createFactory()
const middleware = factory.createMiddleware(async (c, next) => {
c.set('foo', 'bar')
await next()
})
const handlers = factory.createHandlers(logger(), middleware, (c) => {
return c.json(c.var.foo)
})
app.get('/api', ...handlers)Building a larger application
Use app.route() to build a larger application without creating "Ruby on Rails-like Controllers".
If your application has /authors and /books endpoints and you wish to separate files from index.ts, create authors.ts and books.ts.
// authors.ts
import { Hono } from 'hono'
const app = new Hono()
app.get('/', (c) => c.json('list authors'))
app.post('/', (c) => c.json('create an author', 201))
app.get('/:id', (c) => c.json(`get ${c.req.param('id')}`))
export default app// books.ts
import { Hono } from 'hono'
const app = new Hono()
app.get('/', (c) => c.json('list books'))
app.post('/', (c) => c.json('create a book', 201))
app.get('/:id', (c) => c.json(`get ${c.req.param('id')}`))
export default appThen, import them and mount on the paths /authors and /books with app.route().
// index.ts
import { Hono } from 'hono'
import authors from './authors'
import books from './books'
const app = new Hono()
// 😃
app.route('/authors', authors)
app.route('/books', books)
export default appIf you want to use RPC features
The code above works well for normal use cases.
However, if you want to use the RPC feature, you can get the correct type by chaining as follows.
// authors.ts
import { Hono } from 'hono'
const app = new Hono()
.get('/', (c) => c.json('list authors'))
.post('/', (c) => c.json('create an author', 201))
.get('/:id', (c) => c.json(`get ${c.req.param('id')}`))
export default app
export type AppType = typeof appIf you pass the type of the app to hc, it will get the correct type.
import type { AppType } from './authors'
import { hc } from 'hono/client'
// 😃
const client = hc<AppType>('http://localhost') // Typed correctlyFor more detailed information, please see the RPC page.
Guides/Create Hono
Create-hono
Command-line options supported by create-hono - the project initializer that runs when you run npm create hono@latest, npx create-hono@latest, or pnpm create hono@latest.
[!NOTE]
Why this page? The installation / quick-start examples often show a minimalnpm create hono@latest my-appcommand.create-honosupports several useful flags you can pass to automate and customize project creation (select templates, skip prompts, pick a package manager, use local cache, and more).
Passing arguments:
When you use npm create (or npx) arguments intended for the initializer script must be placed after --. Anything after -- is forwarded to the initializer.
::: code-group
# Forwarding arguments to create-hono (npm requires `--`)
npm create hono@latest my-app -- --template cloudflare-workers# "--template cloudflare-workers" selects the Cloudflare Workers template
yarn create hono my-app --template cloudflare-workers# "--template cloudflare-workers" selects the Cloudflare Workers template
pnpm create hono@latest my-app --template cloudflare-workers# "--template cloudflare-workers" selects the Cloudflare Workers template
bun create hono@latest my-app --template cloudflare-workers# "--template cloudflare-workers" selects the Cloudflare Workers template
deno init --npm hono@latest my-app --template cloudflare-workers:::
Commonly used arguments
| Argument | Description | Example |
|---|---|---|
--template <template> |
Select a starter template and skip the interactive template prompt. Templates may include names like bun, cloudflare-workers, vercel, etc. |
--template cloudflare-workers |
--install |
Automatically install dependencies after the template is created. | --install |
--pm |
Specify which package manager to run when installing dependencies. Common values: npm, pnpm, yarn. |
--pm pnpm |
--offline |
Use the local cache/templates instead of fetching the latest remote templates. Useful for offline environments or deterministic local runs. | --offline |
[!NOTE]
The exact set of templates and available options is maintained by thecreate-honoproject. This docs page summarizes the most-used flags — see the linked repository below for the full, authoritative reference.
Example flows
Minimal, interactive
npm create hono@latest my-appThis prompts you for template and options.
Non-interactive, pick template and package manager
npm create hono@latest my-app -- --template vercel --pm npm --installThis creates my-app using the vercel template, installs dependencies using npm, and skips the interactive prompts.
Use offline cache (no network)
pnpm create hono@latest my-app --template deno --offlineTroubleshooting & tips
- If an option appears not to be recognized, make sure you're forwarding it with
--when usingnpm create/npx. - To see the most current list of templates and flags, consult the
create-honorepository or run the initializer locally and follow its help output.
Links & references
create-honorepository : create-hono
Guides/Faq
Frequently Asked Questions
This guide is a collection of frequently asked questions (FAQ) about Hono and how to resolve them.
Is there an official Renovate config for Hono?
The Hono teams does not currently maintain Renovate Configuration.
Therefore, please use third-party renovate-config as follows.
In your renovate.json :
// renovate.json
{
"$schema": "https://docs.renovatebot.com/renovate-schema.json",
"extends": [
"github>shinGangan/renovate-config-hono" // [!code ++]
]
}see renovate-config-hono repository for more details.
Guides/Helpers
Helpers
Helpers are available to assist in developing your application. Unlike middleware, they don't act as handlers, but rather provide useful functions.
For instance, here's how to use the Cookie helper:
import { getCookie, setCookie } from 'hono/cookie'
const app = new Hono()
app.get('/cookie', (c) => {
const yummyCookie = getCookie(c, 'yummy_cookie')
// ...
setCookie(c, 'delicious_cookie', 'macha')
//
})Available Helpers
Guides/Jsx
JSX
You can write HTML with JSX syntax with hono/jsx.
Although hono/jsx works on the client, you will probably use it most often when rendering content on the server side. Here are some things related to JSX that are common to both server and client.
Settings
To use JSX, modify the tsconfig.json:
tsconfig.json:
{
"compilerOptions": {
"jsx": "react-jsx",
"jsxImportSource": "hono/jsx"
}
}Alternatively, use the pragma directives:
/** @jsx jsx */
/** @jsxImportSource hono/jsx */For Deno, you have to modify the deno.json instead of the tsconfig.json:
{
"compilerOptions": {
"jsx": "precompile",
"jsxImportSource": "hono/jsx"
}
}Usage
:::info
If you are coming straight from the Quick Start, the main file has a .ts extension - you need to change it to .tsx - otherwise you will not be able to run the application at all. You should additionally modify the package.json (or deno.json if you are using Deno) to reflect that change (e.g. instead of having bun run --hot src/index.ts in dev script, you should have bun run --hot src/index.tsx).
:::
index.tsx:
import { Hono } from 'hono'
import type { FC } from 'hono/jsx'
const app = new Hono()
const Layout: FC = (props) => {
return (
<html>
{props.children}
</html>
)
}
const Top: FC<{ messages: string[] }> = (props: {
messages: string[]
}) => {
return (
<Layout>
<h1>Hello Hono!</h1>
{props.messages.map((message) => {
return <li>{message}!!</li>
})}
</Layout>
)
}
app.get('/', (c) => {
const messages = ['Good Morning', 'Good Evening', 'Good Night']
return c.html(<Top messages={messages} />)
})
export default appMetadata hoisting
You can write document metadata tags such as <title>, <link>, and <meta> directly inside your components. These tags will be automatically hoisted to the <head> section of the document. This is especially useful when the <head> element is rendered far from the component that determines the appropriate metadata.
import { Hono } from 'hono'
const app = new Hono()
app.use('*', async (c, next) => {
c.setRenderer((content) => {
return c.html(
<html>
<head></head>
{content}
</html>
)
})
await next()
})
app.get('/about', (c) => {
return c.render(
<>
<title>About Page</title>
<meta name='description' content='This is the about page.' />
about page content
</>
)
})
export default appFragment
Use Fragment to group multiple elements without adding extra nodes:
import { Fragment } from 'hono/jsx'
const List = () => (
<Fragment>
first child
second child
third child
</Fragment>
)Or you can write it with <></> if it set up properly.
const List = () => (
<>
first child
second child
third child
</>
)`PropsWithChildren`
You can use PropsWithChildren to correctly infer a child element in a function component.
import { PropsWithChildren } from 'hono/jsx'
type Post = {
id: number
title: string
}
function Component({ title, children }: PropsWithChildren) {
return (
<h1>{title}</h1>
{children}
)
}Inserting Raw HTML
To directly insert HTML, use dangerouslySetInnerHTML:
app.get('/foo', (c) => {
const inner = { __html: 'JSX · SSR' }
const Div =
})Memoization
Optimize your components by memoizing computed strings using memo:
import { memo } from 'hono/jsx'
const Header = memo(() => <header>Welcome to Hono</header>)
const Footer = memo(() => <footer>Powered by Hono</footer>)
const Layout = (
<Header />
Hono is cool!
<Footer />
)Context
By using useContext, you can share data globally across any level of the Component tree without passing values through props.
import type { FC } from 'hono/jsx'
import { createContext, useContext } from 'hono/jsx'
const themes = {
light: {
color: '#000000',
background: '#eeeeee',
},
dark: {
color: '#ffffff',
background: '#222222',
},
}
const ThemeContext = createContext(themes.light)
const Button: FC = () => {
const theme = useContext(ThemeContext)
return Push!
}
const Toolbar: FC = () => {
return (
)
}
// ...
app.get('/', (c) => {
return c.html(
<Toolbar />
)
})Async Component
hono/jsx supports an Async Component, so you can use async/await in your component.
If you render it with c.html(), it will await automatically.
const AsyncComponent = async () => {
await new Promise((r) => setTimeout(r, 1000)) // sleep 1s
return Done!
}
app.get('/', (c) => {
return c.html(
<html>
<AsyncComponent />
</html>
)
})Suspense
The React-like Suspense feature is available.
If you wrap the async component with Suspense, the content in the fallback will be rendered first, and once the Promise is resolved, the awaited content will be displayed.
You can use it with renderToReadableStream().
import { renderToReadableStream, Suspense } from 'hono/jsx/streaming'
//...
app.get('/', (c) => {
const stream = renderToReadableStream(
<html>
<Suspense fallback={loading...}>
<Component />
</Suspense>
</html>
)
return c.body(stream, {
headers: {
'Content-Type': 'text/html; charset=UTF-8',
'Transfer-Encoding': 'chunked',
},
})
})ErrorBoundary
You can catch errors in child components using ErrorBoundary.
In the example below, it will show the content specified in fallback if an error occurs.
function SyncComponent() {
throw new Error('Error')
return Hello
}
app.get('/sync', async (c) => {
return c.html(
<html>
<ErrorBoundary fallback={Out of Service}>
<SyncComponent />
</ErrorBoundary>
</html>
)
})ErrorBoundary can also be used with async components and Suspense.
async function AsyncComponent() {
await new Promise((resolve) => setTimeout(resolve, 2000))
throw new Error('Error')
return Hello
}
app.get('/with-suspense', async (c) => {
return c.html(
<html>
<ErrorBoundary fallback={Out of Service}>
<Suspense fallback={Loading...}>
<AsyncComponent />
</Suspense>
</ErrorBoundary>
</html>
)
})StreamingContext
You can use StreamingContext to provide configuration for streaming components like Suspense and ErrorBoundary. This is useful for adding nonce values to script tags generated by these components for Content Security Policy (CSP).
import { Suspense, StreamingContext } from 'hono/jsx/streaming'
// ...
app.get('/', (c) => {
const stream = renderToReadableStream(
<html>
<StreamingContext
value={{ scriptNonce: 'random-nonce-value' }}
>
<Suspense fallback={Loading...}>
<AsyncComponent />
</Suspense>
</StreamingContext>
</html>
)
return c.body(stream, {
headers: {
'Content-Type': 'text/html; charset=UTF-8',
'Transfer-Encoding': 'chunked',
'Content-Security-Policy':
"script-src 'nonce-random-nonce-value'",
},
})
})The scriptNonce value will be automatically added to any <script> tags generated by Suspense and ErrorBoundary components.
Integration with html Middleware
Combine the JSX and html middlewares for powerful templating.
For in-depth details, consult the html middleware documentation.
import { Hono } from 'hono'
import { html } from 'hono/html'
const app = new Hono()
interface SiteData {
title: string
children?: any
}
const Layout = (props: SiteData) =>
html`<!doctype html>
<html>
<head>
<title>${props.title}</title>
</head>
${props.children}
</html>`
const Content = (props: { siteData: SiteData; name: string }) => (
<Layout {...props.siteData}>
<h1>Hello {props.name}</h1>
</Layout>
)
app.get('/:name', (c) => {
const { name } = c.req.param()
const props = {
name: name,
siteData: {
title: 'JSX with html sample',
},
}
return c.html(<Content {...props} />)
})
export default appWith JSX Renderer Middleware
The JSX Renderer Middleware allows you to create HTML pages more easily with the JSX.
Override type definitions
You can override the type definition to add your custom elements and attributes.
declare module 'hono/jsx' {
namespace JSX {
interface IntrinsicElements {
'my-custom-element': HTMLAttributes & {
'x-event'?: 'click' | 'scroll'
}
}
}
}Guides/Jsx Dom
Client Components
hono/jsx supports not only server side but also client side. This means that it is possible to create an interactive UI that runs in the browser. We call it Client Components or hono/jsx/dom.
It is fast and very small. The counter program in hono/jsx/dom is only 2.8KB with Brotli compression. But, 47.8KB for React.
This section introduces Client Components-specific features.
Counter example
Here is an example of a simple counter, the same code works as in React.
import { useState } from 'hono/jsx'
import { render } from 'hono/jsx/dom'
function Counter() {
const [count, setCount] = useState(0)
return (
Count: {count}
setCount(count + 1)}>Increment
)
}
function App() {
return (
<html>
<Counter />
</html>
)
}
const root = document.getElementById('root')
render(<App />, root)`render()`
You can use render() to insert JSX components within a specified HTML element.
render(<Component />, container)You can see full example code here: Counter example.
Hooks compatible with React
hono/jsx/dom has Hooks that are compatible or partially compatible with React. You can learn about these APIs by looking at the React documentation.
useState()useEffect()useRef()useCallback()use()startTransition()useTransition()useDeferredValue()useMemo()useLayoutEffect()useReducer()useDebugValue()createElement()memo()isValidElement()useId()createRef()forwardRef()useImperativeHandle()useSyncExternalStore()useInsertionEffect()useFormStatus()useActionState()useOptimistic()
`startViewTransition()` family
The startViewTransition() family contains original hooks and functions to handle View Transitions API easily. The followings are examples of how to use them.
1. An easiest example
You can write a transition using the document.startViewTransition shortly with the startViewTransition().
import { useState, startViewTransition } from 'hono/jsx'
import { css, Style } from 'hono/css'
export default function App() {
const [showLargeImage, setShowLargeImage] = useState(false)
return (
<>
<Style />
startViewTransition(() =>
setShowLargeImage((state) => !state)
)
}
>
Click!
{!showLargeImage ? (
) : (
)}
</>
)
}2. Using `viewTransition()` with `keyframes()`
The viewTransition() function allows you to get the unique view-transition-name.
You can use it with the keyframes(), The ::view-transition-old() is converted to ::view-transition-old(${uniqueName)).
import { useState, startViewTransition } from 'hono/jsx'
import { viewTransition } from 'hono/jsx/dom/css'
import { css, keyframes, Style } from 'hono/css'
const rotate = keyframes`
from {
rotate: 0deg;
}
to {
rotate: 360deg;
}
`
export default function App() {
const [showLargeImage, setShowLargeImage] = useState(false)
const [transitionNameClass] = useState(() =>
viewTransition(css`
::view-transition-old() {
animation-name: ${rotate};
}
::view-transition-new() {
animation-name: ${rotate};
}
`)
)
return (
<>
<Style />
startViewTransition(() =>
setShowLargeImage((state) => !state)
)
}
>
Click!
{!showLargeImage ? (
) : (
)}
</>
)
}3. Using `useViewTransition`
If you want to change the style only during the animation. You can use useViewTransition(). This hook returns the [boolean, (callback: () => void) => void], and they are the isUpdating flag and the startViewTransition() function.
When this hook is used, the Component is evaluated at the following two times.
- Inside the callback of a call to
startViewTransition(). - When the
finishpromise becomes fulfilled
import { useState, useViewTransition } from 'hono/jsx'
import { viewTransition } from 'hono/jsx/dom/css'
import { css, keyframes, Style } from 'hono/css'
const rotate = keyframes`
from {
rotate: 0deg;
}
to {
rotate: 360deg;
}
`
export default function App() {
const [isUpdating, startViewTransition] = useViewTransition()
const [showLargeImage, setShowLargeImage] = useState(false)
const [transitionNameClass] = useState(() =>
viewTransition(css`
::view-transition-old() {
animation-name: ${rotate};
}
::view-transition-new() {
animation-name: ${rotate};
}
`)
)
return (
<>
<Style />
startViewTransition(() =>
setShowLargeImage((state) => !state)
)
}
>
Click!
{!showLargeImage ? (
) : (
)}
</>
)
}The `hono/jsx/dom` runtime
There is a small JSX Runtime for Client Components. Using this will result in smaller bundled results than using hono/jsx. Specify hono/jsx/dom in tsconfig.json. For Deno, modify the deno.json.
{
"compilerOptions": {
"jsx": "react-jsx",
"jsxImportSource": "hono/jsx/dom"
}
}Alternatively, you can specify hono/jsx/dom in the esbuild transform options in vite.config.ts.
import { defineConfig } from 'vite'
export default defineConfig({
esbuild: {
jsxImportSource: 'hono/jsx/dom',
},
})Guides/Middleware
Middleware
Middleware works before/after the endpoint Handler. We can get the Request before dispatching or manipulate the Response after dispatching.
Definition of Middleware
- Handler - should return
Responseobject. Only one handler will be called. - Middleware - should
await next()and return nothing to call the next Middleware, or return aResponseto early-exit.
The user can register middleware using app.use or using app.HTTP_METHOD as well as the handlers. For this feature, it's easy to specify the path and the method.
// match any method, all routes
app.use(logger())
// specify path
app.use('/posts/*', cors())
// specify method and path
app.post('/posts/*', basicAuth())If the handler returns Response, it will be used for the end-user, and stopping the processing.
app.post('/posts', (c) => c.text('Created!', 201))In this case, four middleware are processed before dispatching like this:
logger() -> cors() -> basicAuth() -> *handler*Execution order
The order in which Middleware is executed is determined by the order in which it is registered.
The process before the next of the first registered Middleware is executed first,
and the process after the next is executed last.
See below.
app.use(async (_, next) => {
console.log('middleware 1 start')
await next()
console.log('middleware 1 end')
})
app.use(async (_, next) => {
console.log('middleware 2 start')
await next()
console.log('middleware 2 end')
})
app.use(async (_, next) => {
console.log('middleware 3 start')
await next()
console.log('middleware 3 end')
})
app.get('/', (c) => {
console.log('handler')
return c.text('Hello!')
})Result is the following.
middleware 1 start
middleware 2 start
middleware 3 start
handler
middleware 3 end
middleware 2 end
middleware 1 endNote that if the handler or any middleware throws, hono will catch it and either pass it to your app.onError() callback or automatically convert it to a 500 response before returning it up the chain of middleware. This means that next() will never throw, so there is no need to wrap it in a try/catch/finally.
Built-in Middleware
Hono has built-in middleware.
import { Hono } from 'hono'
import { poweredBy } from 'hono/powered-by'
import { logger } from 'hono/logger'
import { basicAuth } from 'hono/basic-auth'
const app = new Hono()
app.use(poweredBy())
app.use(logger())
app.use(
'/auth/*',
basicAuth({
username: 'hono',
password: 'acoolproject',
})
)::: warning
In Deno, it is possible to use a different version of middleware than the Hono version, but this can lead to bugs.
For example, this code is not working because the version is different.
import { Hono } from 'jsr:@hono/[email protected]'
import { upgradeWebSocket } from 'jsr:@hono/[email protected]/deno'
const app = new Hono()
app.get(
'/ws',
upgradeWebSocket(() => ({
// ...
}))
):::
Custom Middleware
You can write your own middleware directly inside app.use():
// Custom logger
app.use(async (c, next) => {
console.log(`[${c.req.method}] ${c.req.url}`)
await next()
})
// Add a custom header
app.use('/message/*', async (c, next) => {
await next()
c.header('x-message', 'This is middleware!')
})
app.get('/message/hello', (c) => c.text('Hello Middleware!'))However, embedding middleware directly within app.use() can limit its reusability. Therefore, we can separate our
middleware into different files.
To ensure we don't lose type definitions for context and next, when separating middleware, we can usecreateMiddleware() from Hono's factory. This also allows us to type-safely access data we've set in Context from downstream handlers.
import { createMiddleware } from 'hono/factory'
const logger = createMiddleware(async (c, next) => {
console.log(`[${c.req.method}] ${c.req.url}`)
await next()
}):::info
Type generics can be used with createMiddleware:
createMiddleware<{Bindings: Bindings}>(async (c, next) =>:::
Modify the Response After Next
Additionally, middleware can be designed to modify responses if necessary:
const stripRes = createMiddleware(async (c, next) => {
await next()
c.res = undefined
c.res = new Response('New Response')
})Context access inside Middleware arguments
To access the context inside middleware arguments, directly use the context parameter provided by app.use. See the example below for clarification.
import { cors } from 'hono/cors'
app.use('*', async (c, next) => {
const middleware = cors({
origin: c.env.CORS_ORIGIN,
})
return middleware(c, next)
})Extending the Context in Middleware
To extend the context inside middleware, use c.set. You can make this type-safe by passing a { Variables: { yourVariable: YourVariableType } } generic argument to the createMiddleware function.
import { createMiddleware } from 'hono/factory'
const echoMiddleware = createMiddleware<{
Variables: {
echo: (str: string) => string
}
}>(async (c, next) => {
c.set('echo', (str) => str)
await next()
})
app.get('/echo', echoMiddleware, (c) => {
return c.text(c.var.echo('Hello!'))
})Third-party Middleware
Built-in middleware does not depend on external modules, but third-party middleware can depend on third-party libraries.
So with them, we may make a more complex application.
We can explore a variety of third-party middleware.
For example, we have GraphQL Server Middleware, Sentry Middleware, Firebase Auth Middleware, and others.
Guides/Others
Miscellaneous
Contributing
Contributions Welcome! You can contribute in the following ways.
- Create an Issue - Propose a new feature. Report a bug.
- Pull Request - Fix a bug and typo. Refactor the code.
- Create third-party middleware - Instruct below.
- Share - Share your thoughts on the Blog, X(Twitter), and others.
- Make your application - Please try to use Hono.
For more details, see Contribution Guide.
Sponsoring
You can sponsor Hono authors via the GitHub sponsor program.
Other Resources
- GitHub repository: https://github.com/honojs
- npm registry: https://www.npmjs.com/package/hono
- JSR: https://jsr.io/@hono/hono
Guides/Rpc
RPC
The RPC feature allows sharing of the API specifications between the server and the client.
First, export the typeof your Hono app (commonly called AppType)—or just the routes you want available to the client—from your server code.
By accepting AppType as a generic parameter, the Hono Client can infer both the input type(s) specified by the Validator, and the output type(s) emitted by handlers returning c.json().
[!NOTE]
For the RPC types to work properly in a monorepo, in both the Client's and Server's tsconfig.json files, set"strict": trueincompilerOptions. Read more.
Server
All you need to do on the server side is to write a validator and create a variable route. The following example uses Zod Validator.
const route = app.post(
'/posts',
zValidator(
'form',
z.object({
title: z.string(),
body: z.string(),
})
),
(c) => {
// ...
return c.json(
{
ok: true,
message: 'Created!',
},
201
)
}
)Then, export the type to share the API spec with the Client.
export type AppType = typeof routeClient
On the Client side, import hc and AppType first.
import type { AppType } from '.'
import { hc } from 'hono/client'hc is a function to create a client. Pass AppType as Generics and specify the server URL as an argument.
const client = hc<AppType>('http://localhost:8787/')Call client.{path}.{method} and pass the data you wish to send to the server as an argument.
const res = await client.posts.$post({
form: {
title: 'Hello',
body: 'Hono is a cool project',
},
})The res is compatible with the "fetch" Response. You can retrieve data from the server with res.json().
if (res.ok) {
const data = await res.json()
console.log(data.message)
}Cookies
To make the client send cookies with every request, add { 'init': { 'credentials": 'include' } } to the options when you're creating the client.
// client.ts
const client = hc<AppType>('http://localhost:8787/', {
init: {
credentials: 'include',
},
})
// This request will now include any cookies you might have set
const res = await client.posts.$get({
query: {
id: '123',
},
})Status code
If you explicitly specify the status code, such as 200 or 404, in c.json(). It will be added as a type for passing to the client.
// server.ts
const app = new Hono().get(
'/posts',
zValidator(
'query',
z.object({
id: z.string(),
})
),
async (c) => {
const { id } = c.req.valid('query')
const post: Post | undefined = await getPost(id)
if (post === undefined) {
return c.json({ error: 'not found' }, 404) // Specify 404
}
return c.json({ post }, 200) // Specify 200
}
)
export type AppType = typeof appYou can get the data by the status code.
// client.ts
const client = hc<AppType>('http://localhost:8787/')
const res = await client.posts.$get({
query: {
id: '123',
},
})
if (res.status === 404) {
const data: { error: string } = await res.json()
console.log(data.error)
}
if (res.ok) {
const data: { post: Post } = await res.json()
console.log(data.post)
}
// { post: Post } | { error: string }
type ResponseType = InferResponseType<typeof client.posts.$get>
// { post: Post }
type ResponseType200 = InferResponseType<
typeof client.posts.$get,
200
>Not Found
If you want to use a client, you should not use c.notFound() for the Not Found response. The data that the client gets from the server cannot be inferred correctly.
// server.ts
export const routes = new Hono().get(
'/posts',
zValidator(
'query',
z.object({
id: z.string(),
})
),
async (c) => {
const { id } = c.req.valid('query')
const post: Post | undefined = await getPost(id)
if (post === undefined) {
return c.notFound() // ❌️
}
return c.json({ post })
}
)
// client.ts
import { hc } from 'hono/client'
const client = hc<typeof routes>('/')
const res = await client.posts[':id'].$get({
param: {
id: '123',
},
})
const data = await res.json() // 🙁 data is unknownPlease use c.json() and specify the status code for the Not Found Response.
export const routes = new Hono().get(
'/posts',
zValidator(
'query',
z.object({
id: z.string(),
})
),
async (c) => {
const { id } = c.req.valid('query')
const post = await getPost(id)
if (!post) {
return c.json({ error: 'not found' }, 404) // Specify 404
}
return c.json({ post }, 200) // Specify 200
}
)Alternatively, you can use module augmentation to extend NotFoundResponse interface. This allows c.notFound() to return a typed response:
// server.ts
import { Hono, TypedResponse } from 'hono'
declare module 'hono' {
interface NotFoundResponse
extends Response,
TypedResponse<{ error: string }, 404, 'json'> {}
}
const app = new Hono()
.get('/posts/:id', async (c) => {
const post = await getPost(c.req.param('id'))
if (!post) {
return c.notFound()
}
return c.json({ post }, 200)
})
.notFound((c) => c.json({ error: 'not found' }, 404))
export type AppType = typeof appNow the client can correctly infer the 404 response type.
Path parameters
You can also handle routes that include path parameters or query values.
const route = app.get(
'/posts/:id',
zValidator(
'query',
z.object({
page: z.coerce.number().optional(), // coerce to convert to number
})
),
(c) => {
// ...
return c.json({
title: 'Night',
body: 'Time to sleep',
})
}
)Both path parameters and query values must be passed as string, even if the underlying value is of a different type.
Specify the string you want to include in the path with param, and any query values with query.
const res = await client.posts[':id'].$get({
param: {
id: '123',
},
query: {
page: '1', // `string`, converted by the validator to `number`
},
})Multiple parameters
Handle routes with multiple parameters.
const route = app.get(
'/posts/:postId/:authorId',
zValidator(
'query',
z.object({
page: z.string().optional(),
})
),
(c) => {
// ...
return c.json({
title: 'Night',
body: 'Time to sleep',
})
}
)Add multiple [''] to specify params in path.
const res = await client.posts[':postId'][':authorId'].$get({
param: {
postId: '123',
authorId: '456',
},
query: {},
})Include slashes
hc function does not URL-encode the values of param. To include slashes in parameters, use regular expressions.
// client.ts
// Requests /posts/123/456
const res = await client.posts[':id'].$get({
param: {
id: '123/456',
},
})
// server.ts
const route = app.get(
'/posts/:id{.+}',
zValidator(
'param',
z.object({
id: z.string(),
})
),
(c) => {
// id: 123/456
const { id } = c.req.valid('param')
// ...
}
)[!NOTE]
Basic path parameters without regular expressions do not match slashes. If you pass aparamcontaining slashes using the hc function, the server might not route as intended. Encoding the parameters usingencodeURIComponentis the recommended approach to ensure correct routing.
Headers
You can append the headers to the request.
const res = await client.search.$get(
{
//...
},
{
headers: {
'X-Custom-Header': 'Here is Hono Client',
'X-User-Agent': 'hc',
},
}
)To add a common header to all requests, specify it as an argument to the hc function.
const client = hc<AppType>('/api', {
headers: {
Authorization: 'Bearer TOKEN',
},
})`init` option
You can pass the fetch's RequestInit object to the request as the init option. Below is an example of aborting a Request.
import { hc } from 'hono/client'
const client = hc<AppType>('http://localhost:8787/')
const abortController = new AbortController()
const res = await client.api.posts.$post(
{
json: {
// Request body
},
},
{
// RequestInit object
init: {
signal: abortController.signal,
},
}
)
// ...
abortController.abort()::: info
A RequestInit object defined by init takes the highest priority. It could be used to overwrite things set by other options like body | method | headers.
:::
`$url()`
You can get a URL object for accessing the endpoint by using $url().
::: warning
You have to pass in an absolute URL for this to work. Passing in a relative URL / will result in the following error.
Uncaught TypeError: Failed to construct 'URL': Invalid URL
// ❌ Will throw error
const client = hc<AppType>('/')
client.api.post.$url()
// ✅ Will work as expected
const client = hc<AppType>('http://localhost:8787/')
client.api.post.$url():::
const route = app
.get('/api/posts', (c) => c.json({ posts }))
.get('/api/posts/:id', (c) => c.json({ post }))
const client = hc<typeof route>('http://localhost:8787/')
let url = client.api.posts.$url()
console.log(url.pathname) // `/api/posts`
url = client.api.posts[':id'].$url({
param: {
id: '123',
},
})
console.log(url.pathname) // `/api/posts/123`Typed URL
You can pass the base URL as the second type parameter to hc to get more precise URL types:
const client = hc<typeof route, 'http://localhost:8787'>(
'http://localhost:8787/'
)
const url = client.api.posts.$url()
// url is TypedURL with precise type information
// including protocol, host, and pathThis is useful when you want to use the URL as a type-safe key for libraries like SWR.
File Uploads
You can upload files using a form body:
// client
const res = await client.user.picture.$put({
form: {
file: new File([fileToUpload], filename, {
type: fileToUpload.type,
}),
},
})// server
const route = app.put(
'/user/picture',
zValidator(
'form',
z.object({
file: z.instanceof(File),
})
)
// ...
)Custom `fetch` method
You can set the custom fetch method.
In the following example script for Cloudflare Worker, the Service Bindings' fetch method is used instead of the default fetch.
# wrangler.toml
services = [
{ binding = "AUTH", service = "auth-service" },
]// src/client.ts
const client = hc<CreateProfileType>('http://localhost', {
fetch: c.env.AUTH.fetch.bind(c.env.AUTH),
})Custom query serializer
You can customize how query parameters are serialized using the buildSearchParams option. This is useful when you need bracket notation for arrays or other custom formats:
const client = hc<AppType>('http://localhost', {
buildSearchParams: (query) => {
const searchParams = new URLSearchParams()
for (const [k, v] of Object.entries(query)) {
if (v === undefined) {
continue
}
if (Array.isArray(v)) {
v.forEach((item) => searchParams.append(`${k}[]`, item))
} else {
searchParams.set(k, v)
}
}
return searchParams
},
})Infer
Use InferRequestType and InferResponseType to know the type of object to be requested and the type of object to be returned.
import type { InferRequestType, InferResponseType } from 'hono/client'
// InferRequestType
const $post = client.todo.$post
type ReqType = InferRequestType<typeof $post>['form']
// InferResponseType
type ResType = InferResponseType<typeof $post>Parsing a Response with type-safety helper
You can use parseResponse() helper to easily parse a Response from hc with type-safety.
import { parseResponse, DetailedError } from 'hono/client'
// result contains the parsed response body (automatically parsed based on Content-Type)
const result = await parseResponse(client.hello.$get()).catch(
(e: DetailedError) => {
console.error(e)
}
)
// parseResponse automatically throws an error if response is not okUsing SWR
You can also use a React Hook library such as SWR.
import useSWR from 'swr'
import { hc } from 'hono/client'
import type { InferRequestType } from 'hono/client'
import type { AppType } from '../functions/api/[[route]]'
const App = () => {
const client = hc<AppType>('/api')
const $get = client.hello.$get
const fetcher =
(arg: InferRequestType<typeof $get>) => async () => {
const res = await $get(arg)
return await res.json()
}
const { data, error, isLoading } = useSWR(
'api-hello',
fetcher({
query: {
name: 'SWR',
},
})
)
if (error) return failed to load
if (isLoading) return loading...
return <h1>{data?.message}</h1>
}
export default AppUsing RPC with larger applications
In the case of a larger application, such as the example mentioned in Building a larger application, you need to be careful about the type of inference.
A simple way to do this is to chain the handlers so that the types are always inferred.
// authors.ts
import { Hono } from 'hono'
const app = new Hono()
.get('/', (c) => c.json('list authors'))
.post('/', (c) => c.json('create an author', 201))
.get('/:id', (c) => c.json(`get ${c.req.param('id')}`))
export default app// books.ts
import { Hono } from 'hono'
const app = new Hono()
.get('/', (c) => c.json('list books'))
.post('/', (c) => c.json('create a book', 201))
.get('/:id', (c) => c.json(`get ${c.req.param('id')}`))
export default appYou can then import the sub-routers as you usually would, and make sure you chain their handlers as well, since this is the top level of the app in this case, this is the type we'll want to export.
// index.ts
import { Hono } from 'hono'
import authors from './authors'
import books from './books'
const app = new Hono()
const routes = app.route('/authors', authors).route('/books', books)
export default app
export type AppType = typeof routesYou can now create a new client using the registered AppType and use it as you would normally.
Known issues
IDE performance
When using RPC, the more routes you have, the slower your IDE will become. One of the main reasons for this is that massive amounts of type instantiations are executed to infer the type of your app.
For example, suppose your app has a route like this:
// app.ts
export const app = new Hono().get('foo/:id', (c) =>
c.json({ ok: true }, 200)
)Hono will infer the type as follows:
export const app = Hono().get<
'foo/:id',
'foo/:id',
JSONRespondReturn<{ ok: boolean }, 200>,
BlankInput,
BlankEnv
>('foo/:id', (c) => c.json({ ok: true }, 200))This is a type instantiation for a single route. While the user doesn't need to write these type arguments manually, which is a good thing, it's known that type instantiation takes much time. tsserver used in your IDE does this time consuming task every time you use the app. If you have a lot of routes, this can slow down your IDE significantly.
However, we have some tips to mitigate this issue.
Hono version mismatch
If your backend is separate from the frontend and lives in a different directory, you need to ensure that the Hono versions match. If you use one Hono version on the backend and another on the frontend, you'll run into issues such as "Type instantiation is excessively deep and possibly infinite".
TypeScript project references
Like in the case of Hono version mismatch, you'll run into issues if your backend and frontend are separate. If you want to access code from the backend (AppType, for example) on the frontend, you need to use project references. TypeScript's project references allow one TypeScript codebase to access and use code from another TypeScript codebase. (source: Hono RPC And TypeScript Project References).
Compile your code before using it (recommended)
tsc can do heavy tasks like type instantiation at compile time! Then, tsserver doesn't need to instantiate all the type arguments every time you use it. It will make your IDE a lot faster!
Compiling your client including the server app gives you the best performance. Put the following code in your project:
import { app } from './app'
import { hc } from 'hono/client'
// this is a trick to calculate the type when compiling
export type Client = ReturnType<typeof hc<typeof app>>
export const hcWithType = (...args: Parameters<typeof hc>): Client =>
hc<typeof app>(...args)After compiling, you can use hcWithType instead of hc to get the client with the type already calculated.
const client = hcWithType('http://localhost:8787/')
const res = await client.posts.$post({
form: {
title: 'Hello',
body: 'Hono is a cool project',
},
})If your project is a monorepo, this solution does fit well. Using a tool like turborepo, you can easily separate the server project and the client project and get better integration managing dependencies between them. Here is a working example.
You can also coordinate your build process manually with tools like concurrently or npm-run-all.
Specify type arguments manually
This is a bit cumbersome, but you can specify type arguments manually to avoid type instantiation.
const app = new Hono().get<'foo/:id'>('foo/:id', (c) =>
c.json({ ok: true }, 200)
)Specifying just single type argument make a difference in performance, while it may take you a lot of time and effort if you have a lot of routes.
Split your app and client into multiple files
As described in Using RPC with larger applications, you can split your app into multiple apps. You can also create a client for each app:
// authors-cli.ts
import { app as authorsApp } from './authors'
import { hc } from 'hono/client'
const authorsClient = hc<typeof authorsApp>('/authors')
// books-cli.ts
import { app as booksApp } from './books'
import { hc } from 'hono/client'
const booksClient = hc<typeof booksApp>('/books')This way, tsserver doesn't need to instantiate types for all routes at once.
Guides/Testing
Testing
Testing is important.
In actuality, it is easy to test Hono's applications.
The way to create a test environment differs from each runtime, but the basic steps are the same.
In this section, let's test with Cloudflare Workers and Vitest.
::: tip
Cloudflare recommends using Vitest with @cloudflare/vitest-pool-workers. For more details, please refer to Vitest integration in the Cloudflare Workers docs.
:::
Request and Response
All you do is create a Request and pass it to the Hono application to validate the Response.
And, you can use app.request the useful method.
::: tip
For a typed test client see the testing helper.
:::
For example, consider an application that provides the following REST API.
app.get('/posts', (c) => {
return c.text('Many posts')
})
app.post('/posts', (c) => {
return c.json(
{
message: 'Created',
},
201,
{
'X-Custom': 'Thank you',
}
)
})Make a request to GET /posts and test the response.
describe('Example', () => {
test('GET /posts', async () => {
const res = await app.request('/posts')
expect(res.status).toBe(200)
expect(await res.text()).toBe('Many posts')
})
})To make a request to POST /posts, do the following.
test('POST /posts', async () => {
const res = await app.request('/posts', {
method: 'POST',
})
expect(res.status).toBe(201)
expect(res.headers.get('X-Custom')).toBe('Thank you')
expect(await res.json()).toEqual({
message: 'Created',
})
})To make a request to POST /posts with JSON data, do the following.
test('POST /posts', async () => {
const res = await app.request('/posts', {
method: 'POST',
body: JSON.stringify({ message: 'hello hono' }),
headers: new Headers({ 'Content-Type': 'application/json' }),
})
expect(res.status).toBe(201)
expect(res.headers.get('X-Custom')).toBe('Thank you')
expect(await res.json()).toEqual({
message: 'Created',
})
})To make a request to POST /posts with multipart/form-data data, do the following.
test('POST /posts', async () => {
const formData = new FormData()
formData.append('message', 'hello')
const res = await app.request('/posts', {
method: 'POST',
body: formData,
})
expect(res.status).toBe(201)
expect(res.headers.get('X-Custom')).toBe('Thank you')
expect(await res.json()).toEqual({
message: 'Created',
})
})You can also pass an instance of the Request class.
test('POST /posts', async () => {
const req = new Request('http://localhost/posts', {
method: 'POST',
})
const res = await app.request(req)
expect(res.status).toBe(201)
expect(res.headers.get('X-Custom')).toBe('Thank you')
expect(await res.json()).toEqual({
message: 'Created',
})
})In this way, you can test it as like an End-to-End.
Env
To set c.env for testing, you can pass it as the 3rd parameter to app.request. This is useful for mocking values like Cloudflare Workers Bindings:
const MOCK_ENV = {
API_HOST: 'example.com',
DB: {
prepare: () => {
/* mocked D1 */
},
},
}
test('GET /posts', async () => {
const res = await app.request('/posts', {}, MOCK_ENV)
})Guides/Validation
Validation
Hono provides only a very thin Validator.
But, it can be powerful when combined with a third-party Validator.
In addition, the RPC feature allows you to share API specifications with your clients through types.
Manual validator
First, introduce a way to validate incoming values without using the third-party Validator.
Import validator from hono/validator.
import { validator } from 'hono/validator'To validate form data, specify form as the first argument and a callback as the second argument.
In the callback, validates the value and return the validated values at the end.
The validator can be used as middleware.
app.post(
'/posts',
validator('form', (value, c) => {
const body = value['body']
if (!body || typeof body !== 'string') {
return c.text('Invalid!', 400)
}
return {
body: body,
}
}),
//...Within the handler you can get the validated value with c.req.valid('form').
, (c) => {
const { body } = c.req.valid('form')
// ... do something
return c.json(
{
message: 'Created!',
},
201
)
}Validation targets include json, query, header, param and cookie in addition to form.
::: warning
When you validate json or form, the request must contain a matching content-type header (e.g. Content-Type: application/json for json). Otherwise, the request body will not be parsed and you will receive an empty object ({}) as value in the callback.
It is important to set the content-type header when testing usingapp.request().
Given an application like this.
const app = new Hono()
app.post(
'/testing',
validator('json', (value, c) => {
// pass-through validator
return value
}),
(c) => {
const body = c.req.valid('json')
return c.json(body)
}
)Your tests can be written like this.
// ❌ this will not work
const res = await app.request('/testing', {
method: 'POST',
body: JSON.stringify({ key: 'value' }),
})
const data = await res.json()
console.log(data) // {}
// ✅ this will work
const res = await app.request('/testing', {
method: 'POST',
body: JSON.stringify({ key: 'value' }),
headers: new Headers({ 'Content-Type': 'application/json' }),
})
const data = await res.json()
console.log(data) // { key: 'value' }:::
::: warning
When you validate header, you need to use lowercase name as the key.
If you want to validate the Idempotency-Key header, you need to use idempotency-key as the key.
// ❌ this will not work
app.post(
'/api',
validator('header', (value, c) => {
// idempotencyKey is always undefined
// so this middleware always return 400 as not expected
const idempotencyKey = value['Idempotency-Key']
if (idempotencyKey == undefined || idempotencyKey === '') {
throw new HTTPException(400, {
message: 'Idempotency-Key is required',
})
}
return { idempotencyKey }
}),
(c) => {
const { idempotencyKey } = c.req.valid('header')
// ...
}
)
// ✅ this will work
app.post(
'/api',
validator('header', (value, c) => {
// can retrieve the value of the header as expected
const idempotencyKey = value['idempotency-key']
if (idempotencyKey == undefined || idempotencyKey === '') {
throw new HTTPException(400, {
message: 'Idempotency-Key is required',
})
}
return { idempotencyKey }
}),
(c) => {
const { idempotencyKey } = c.req.valid('header')
// ...
}
):::
Multiple validators
You can also include multiple validators to validate different parts of request:
app.post(
'/posts/:id',
validator('param', ...),
validator('query', ...),
validator('json', ...),
(c) => {
//...
}With Zod
You can use Zod, one of third-party validators.
We recommend using a third-party validator.
Install from the Npm registry.
::: code-group
npm i zodyarn add zodpnpm add zodbun add zod:::
Import z from zod.
import * as z from 'zod'Write your schema.
const schema = z.object({
body: z.string(),
})You can use the schema in the callback function for validation and return the validated value.
const route = app.post(
'/posts',
validator('form', (value, c) => {
const parsed = schema.safeParse(value)
if (!parsed.success) {
return c.text('Invalid!', 401)
}
return parsed.data
}),
(c) => {
const { body } = c.req.valid('form')
// ... do something
return c.json(
{
message: 'Created!',
},
201
)
}
)Zod Validator Middleware
You can use the Zod Validator Middleware to make it even easier.
::: code-group
npm i @hono/zod-validatoryarn add @hono/zod-validatorpnpm add @hono/zod-validatorbun add @hono/zod-validator:::
And import zValidator.
import { zValidator } from '@hono/zod-validator'And write as follows.
const route = app.post(
'/posts',
zValidator(
'form',
z.object({
body: z.string(),
})
),
(c) => {
const validated = c.req.valid('form')
// ... use your validated data
}
)Standard Schema Validator Middleware
Standard Schema is a specification that provides a common interface for TypeScript validation libraries. It was created by the maintainers of Zod, Valibot, and ArkType to allow ecosystem tools to work with any validation library without needing custom adapters.
The Standard Schema Validator Middleware lets you use any Standard Schema-compatible validation library with Hono, giving you the flexibility to choose your preferred validator while maintaining consistent type safety.
::: code-group
npm i @hono/standard-validatoryarn add @hono/standard-validatorpnpm add @hono/standard-validatorbun add @hono/standard-validator:::
Import sValidator from the package:
import { sValidator } from '@hono/standard-validator'With Zod
You can use Zod with the Standard Schema validator:
::: code-group
npm i zodyarn add zodpnpm add zodbun add zod:::
import { z } from 'zod'
import { sValidator } from '@hono/standard-validator'
const schema = z.object({
name: z.string(),
age: z.number(),
})
app.post('/author', sValidator('json', schema), (c) => {
const data = c.req.valid('json')
return c.json({
success: true,
message: `${data.name} is ${data.age}`,
})
})With Valibot
Valibot is a lightweight alternative to Zod with a modular design:
::: code-group
npm i valibotyarn add valibotpnpm add valibotbun add valibot:::
import * as v from 'valibot'
import { sValidator } from '@hono/standard-validator'
const schema = v.object({
name: v.string(),
age: v.number(),
})
app.post('/author', sValidator('json', schema), (c) => {
const data = c.req.valid('json')
return c.json({
success: true,
message: `${data.name} is ${data.age}`,
})
})With ArkType
ArkType offers TypeScript-native syntax for runtime validation:
::: code-group
npm i arktypeyarn add arktypepnpm add arktypebun add arktype:::
import { type } from 'arktype'
import { sValidator } from '@hono/standard-validator'
const schema = type({
name: 'string',
age: 'number',
})
app.post('/author', sValidator('json', schema), (c) => {
const data = c.req.valid('json')
return c.json({
success: true,
message: `${data.name} is ${data.age}`,
})
})Getting Started/Ali Function Compute
Alibaba Cloud Function Compute
Alibaba Cloud Function Compute is a fully managed, event-driven compute service. Function Compute allows you to focus on writing and uploading code without having to manage infrastructure such as servers.
This guide uses a third-party adapter rwv/hono-alibaba-cloud-fc3-adapter to run Hono on Alibaba Cloud Function Compute.
1. Setup
::: code-group
mkdir my-app
cd my-app
npm i hono hono-alibaba-cloud-fc3-adapter
npm i -D @serverless-devs/s esbuild
mkdir src
touch src/index.tsmkdir my-app
cd my-app
yarn add hono hono-alibaba-cloud-fc3-adapter
yarn add -D @serverless-devs/s esbuild
mkdir src
touch src/index.tsmkdir my-app
cd my-app
pnpm add hono hono-alibaba-cloud-fc3-adapter
pnpm add -D @serverless-devs/s esbuild
mkdir src
touch src/index.tsmkdir my-app
cd my-app
bun add hono hono-alibaba-cloud-fc3-adapter
bun add -D esbuild @serverless-devs/s
mkdir src
touch src/index.ts:::
2. Hello World
Edit src/index.ts.
import { Hono } from 'hono'
import { handle } from 'hono-alibaba-cloud-fc3-adapter'
const app = new Hono()
app.get('/', (c) => c.text('Hello Hono!'))
export const handler = handle(app)3. Setup serverless-devs
serverless-devs is an open source and open serverless developer platform dedicated to providing developers with a powerful tool chain system. Through this platform, developers can not only experience multi cloud serverless products with one click and rapidly deploy serverless projects, but also manage projects in the whole life cycle of serverless applications, and combine serverless devs with other tools / platforms very simply and quickly to further improve the efficiency of R & D, operation and maintenance.
Add the Alibaba Cloud AccessKeyID & AccessKeySecret
npx s config add
# Please select a provider: Alibaba Cloud (alibaba)
# Input your AccessKeyID & AccessKeySecretEdit s.yaml
edition: 3.0.0
name: my-app
access: 'default'
vars:
region: 'us-west-1'
resources:
my-app:
component: fc3
props:
region: ${vars.region}
functionName: 'my-app'
runtime: 'nodejs20'
code: ./dist
handler: index.handler
memorySize: 1024
timeout: 300Edit scripts section in package.json:
{
"scripts": {
"build": "esbuild --bundle --outfile=./dist/index.js --platform=node --target=node20 ./src/index.ts",
"deploy": "s deploy -y"
}
}4. Deploy
Finally, run the command to deploy:
npm run build # Compile the TypeScript code to JavaScript
npm run deploy # Deploy the function to Alibaba Cloud Function ComputeGetting Started/Aws Lambda
AWS Lambda
AWS Lambda is a serverless platform by Amazon Web Services.
You can run your code in response to events and automatically manages the underlying compute resources for you.
Hono works on AWS Lambda with the Node.js 18+ environment.
1. Setup
When creating the application on AWS Lambda,
CDK
is useful to setup the functions such as IAM Role, API Gateway, and others.
Initialize your project with the cdk CLI.
::: code-group
mkdir my-app
cd my-app
cdk init app -l typescript
npm i hono
npm i -D esbuild
mkdir lambda
touch lambda/index.tsmkdir my-app
cd my-app
cdk init app -l typescript
yarn add hono
yarn add -D esbuild
mkdir lambda
touch lambda/index.tsmkdir my-app
cd my-app
cdk init app -l typescript
pnpm add hono
pnpm add -D esbuild
mkdir lambda
touch lambda/index.tsmkdir my-app
cd my-app
cdk init app -l typescript
bun add hono
bun add -D esbuild
mkdir lambda
touch lambda/index.ts:::
2. Hello World
Edit lambda/index.ts.
import { Hono } from 'hono'
import { handle } from 'hono/aws-lambda'
const app = new Hono()
app.get('/', (c) => c.text('Hello Hono!'))
export const handler = handle(app)3. Deploy
Edit lib/my-app-stack.ts.
import * as cdk from 'aws-cdk-lib'
import { Construct } from 'constructs'
import * as lambda from 'aws-cdk-lib/aws-lambda'
import { NodejsFunction } from 'aws-cdk-lib/aws-lambda-nodejs'
export class MyAppStack extends cdk.Stack {
constructor(scope: Construct, id: string, props?: cdk.StackProps) {
super(scope, id, props)
const fn = new NodejsFunction(this, 'lambda', {
entry: 'lambda/index.ts',
handler: 'handler',
runtime: lambda.Runtime.NODEJS_22_X,
})
const fnUrl = fn.addFunctionUrl({
authType: lambda.FunctionUrlAuthType.NONE,
})
new cdk.CfnOutput(this, 'lambdaUrl', {
value: fnUrl.url!,
})
}
}Finally, run the command to deploy:
cdk deployServe Binary data
Hono supports binary data as a response.
In Lambda, base64 encoding is required to return binary data.
Once binary type is set to Content-Type header, Hono automatically encodes data to base64.
app.get('/binary', async (c) => {
// ...
c.status(200)
c.header('Content-Type', 'image/png') // means binary data
return c.body(buffer) // supports `ArrayBufferLike` type, encoded to base64.
})Access AWS Lambda Object
In Hono, you can access the AWS Lambda Events and Context by binding the LambdaEvent, LambdaContext type and using c.env
import { Hono } from 'hono'
import type { LambdaEvent, LambdaContext } from 'hono/aws-lambda'
import { handle } from 'hono/aws-lambda'
type Bindings = {
event: LambdaEvent
lambdaContext: LambdaContext
}
const app = new Hono<{ Bindings: Bindings }>()
app.get('/aws-lambda-info/', (c) => {
return c.json({
isBase64Encoded: c.env.event.isBase64Encoded,
awsRequestId: c.env.lambdaContext.awsRequestId,
})
})
export const handler = handle(app)Access RequestContext
In Hono, you can access the AWS Lambda request context by binding the LambdaEvent type and using c.env.event.requestContext.
import { Hono } from 'hono'
import type { LambdaEvent } from 'hono/aws-lambda'
import { handle } from 'hono/aws-lambda'
type Bindings = {
event: LambdaEvent
}
const app = new Hono<{ Bindings: Bindings }>()
app.get('/custom-context/', (c) => {
const lambdaContext = c.env.event.requestContext
return c.json(lambdaContext)
})
export const handler = handle(app)Before v3.10.0 (deprecated)
you can access the AWS Lambda request context by binding the ApiGatewayRequestContext type and using c.env.
import { Hono } from 'hono'
import type { ApiGatewayRequestContext } from 'hono/aws-lambda'
import { handle } from 'hono/aws-lambda'
type Bindings = {
requestContext: ApiGatewayRequestContext
}
const app = new Hono<{ Bindings: Bindings }>()
app.get('/custom-context/', (c) => {
const lambdaContext = c.env.requestContext
return c.json(lambdaContext)
})
export const handler = handle(app)Lambda response streaming
By changing the invocation mode of AWS Lambda, you can achieve Streaming Response.
fn.addFunctionUrl({
authType: lambda.FunctionUrlAuthType.NONE,
+ invokeMode: lambda.InvokeMode.RESPONSE_STREAM,
})Typically, the implementation requires writing chunks to NodeJS.WritableStream using awslambda.streamifyResponse, but with the AWS Lambda Adaptor, you can achieve the traditional streaming response of Hono by using streamHandle instead of handle.
import { Hono } from 'hono'
import { streamHandle } from 'hono/aws-lambda'
import { streamText } from 'hono/streaming'
const app = new Hono()
app.get('/stream', async (c) => {
return streamText(c, async (stream) => {
for (let i = 0; i < 3; i++) {
await stream.writeln(`${i}`)
await stream.sleep(1)
}
})
})
export const handler = streamHandle(app)Getting Started/Azure Functions
Azure Functions
Azure Functions is a serverless platform from Microsoft Azure. You can run your code in response to events, and it automatically manages the underlying compute resources for you.
Hono was not designed for Azure Functions at first. But with Azure Functions Adapter it can run on it as well.
It works with Azure Functions V4 running on Node.js 18 or above.
1. Install CLI
To create an Azure Function, you must first install Azure Functions Core Tools.
On macOS
brew tap azure/functions
brew install azure-functions-core-tools@4Follow this link for other OS:
2. Setup
Create a TypeScript Node.js V4 project in the current folder.
func init --typescriptChange the default route prefix of the host. Add this property to the root json object of host.json:
"extensions": {
"http": {
"routePrefix": ""
}
}::: info
The default Azure Functions route prefix is /api. If you don't change it as shown above, be sure to start all your Hono routes with /api
:::
Now you are ready to install Hono and the Azure Functions Adapter with:
::: code-group
npm i @marplex/hono-azurefunc-adapter honoyarn add @marplex/hono-azurefunc-adapter honopnpm add @marplex/hono-azurefunc-adapter honobun add @marplex/hono-azurefunc-adapter hono:::
3. Hello World
Create src/app.ts:
// src/app.ts
import { Hono } from 'hono'
const app = new Hono()
app.get('/', (c) => c.text('Hello Azure Functions!'))
export default appCreate src/functions/httpTrigger.ts:
// src/functions/httpTrigger.ts
import { app } from '@azure/functions'
import { azureHonoHandler } from '@marplex/hono-azurefunc-adapter'
import honoApp from '../app'
app.http('httpTrigger', {
methods: [
//Add all your supported HTTP methods here
'GET',
'POST',
'DELETE',
'PUT',
],
authLevel: 'anonymous',
route: '{*proxy}',
handler: azureHonoHandler(honoApp.fetch),
})4. Run
Run the development server locally. Then, access http://localhost:7071 in your Web browser.
::: code-group
npm run startyarn startpnpm startbun run start:::
5. Deploy
::: info
Before you can deploy to Azure, you need to create some resources in your cloud infrastructure. Please visit the Microsoft documentation on Create supporting Azure resources for your function
:::
Build the project for deployment:
::: code-group
npm run buildyarn buildpnpm buildbun run build:::
Deploy your project to the function app in Azure Cloud. Replace <YourFunctionAppName> with the name of your app.
func azure functionapp publish <YourFunctionAppName>