$ npm install typeboximport Type from 'typebox'
const T = Type.Object({ // const T = {
x: Type.Number(), // type: 'object',
y: Type.Number(), // properties: {
z: Type.Number() // x: { type: 'number' },
}) // y: { type: 'number' },
// z: { type: 'number' }
// },
// required: ['x', 'y', 'z']
// }
type T = Type.Static<typeof T> // type T = {
// x: number,
// y: number,
// z: number
// }TypeBox is a runtime type system that creates in-memory JSON Schema objects that infer as TypeScript types. The schematics produced by this library are designed to match the static type checking rules of the TypeScript compiler. TypeBox offers a unified type that can be statically checked by TypeScript and validated at runtime using standard JSON Schema.
This library is designed to allow JSON Schema to compose similar to how types compose within TypeScript's type system. It can be used as a simple tool to build up complex schematics or integrated into REST and RPC services to help validate data received over the wire.
License: MIT
TypeBox types are JSON Schema fragments that compose into more complex types. The library offers a set of types used to construct JSON Schema compliant schematics as well as a set of extended types used to model constructs native to the JavaScript language. The schematics produced by TypeBox can be passed directly to any JSON Schema compliant validator.
The following creates a User type and infers with Static.
import Type from 'typebox'
// Type
const User = Type.Object({ // const User = {
id: Type.String(), // type: 'object',
name: Type.String(), // properties: {
email: Type.String({ format: 'email' }) // id: { type: 'string' },
}) // name: { type: 'string' },
// email: {
// type: 'string',
// format: 'email'
// }
// }
// required: [
// 'id',
// 'name',
// 'email'
// ]
// }
// Static
type User = Type.Static<typeof User> // type User = {
// id: string,
// name: string,
// email: string
// }Documentation | Example 1 | Example 2
TypeBox includes a micro DSL for composing JSON Schema with TypeScript syntax. The DSL offers a full syntactic frontend to Type.* and supports many advanced type-level constructs such as Conditional, Mapped, Indexed, Infer, Generics, Distributed types and more. This feature is implemented symmetrically at runtime and statically via TypeScript Template Literal types.
The following uses Script to parse a TypeScript definition module into JSON Schema.
import Type from 'typebox'
// Script
const { User, UserUpdate } = Type.Script(`
interface Entity {
id: string
}
interface User extends Entity {
name: string,
email: string
}
type UserUpdate = Evaluate<
Pick<User, keyof Entity> &
Partial<Omit<User, keyof Entity>>
>
`)
// Reflect
console.log(User) // {
// type: 'object',
// properties: {
// id: { type: 'string' },
// name: { type: 'string' },
// email: { type: 'string' }
// },
// required: [
// 'id',
// 'name',
// 'email'
// ]
// }
console.log(UserUpdate) // {
// type: 'object',
// properties: {
// id: { type: 'string' },
// name: { type: 'string' },
// email: { type: 'string' }
// },
// required: ['id']
// }
// Static
type User = Type.Static<typeof User> // type User = {
// id: string,
// name: string,
// email: string
// }
type UserUpdate = Type.Static<typeof UserUpdate> // type UserUpdate = {
// id: string,
// name?: string,
// email?: string
// }
Documentation | Example 1 | Example 2
TypeBox includes a high-performance JIT compiler that supports JSON Schema Draft 3 through to 2020-12. It is designed to be a lightweight industry-grade alternative to Ajv and offers improved compilation and validation performance. It also provides automatic fallback to dynamic validation in JIT restricted environments such as Cloudflare Workers.
The compiler is available via optional sub module import.
import Schema from 'typebox/schema'The compiler accepts JSON Schema and returns Validator instances.
const Vector = Schema.Compile(Type.Object({ // const Vector: Validator<TObject<{
x: Type.Number(), // x: TNumber
y: Type.Number(), // y: TNumber
z: Type.Number() // z: TNumber
})) // }>>With JSON Schema
const Vector = Schema.Compile({ // const Vector: Validator<{
type: 'object', // type: "object";
required: ['x', 'y', 'z'], // required: ["x", "y", "z"];
properties: { // properties: { ... };
x: { type: 'number' }, // }, { ... }>
y: { type: 'number' },
z: { type: 'number' }
}
})Validator instances provide functions to Check and Parse values.
const Vector = Schema.Compile(Type.Script(`{
x: number
y: number
z: number
}`))
const valid = Vector.Check({ // const valid: boolean
x: 1,
y: 0,
z: 0
})
const value = Vector.Parse({ // const value: {
x: 1, // x: number
y: 0, // y: number
z: 0 // z: number
}) // }The following table shows specification coverage implemented by TypeBox.
| Spec | 3 | 4 | 6 | 7 | 2019-09 | 2020-12 | v1 |
|---|---|---|---|---|---|---|---|
| additionalItems | ✅ | ✅ | ✅ | ✅ | ✅ | - | - |
| additionalProperties | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| allOf | - | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| anchor | - | - | - | - | ✅ | ✅ | ✅ |
| anyOf | - | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| boolean_schema | - | - | ✅ | ✅ | ✅ | ✅ | ✅ |
| const | - | - | ✅ | ✅ | ✅ | ✅ | ✅ |
| contains | - | - | ✅ | ✅ | ✅ | ✅ | ✅ |
| content | - | - | - | - | ✅ | ✅ | ✅ |
| default | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| dependencies | 17/18 | ✅ | ✅ | ✅ | - | - | - |
| dependentRequired | - | - | - | - | ✅ | ✅ | ✅ |
| dependentSchemas | - | - | - | - | ✅ | ✅ | ✅ |
| dynamicRef | - | - | - | - | - | 38/44 | 19/27 |
| enum | 14/16 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| exclusiveMaximum | - | - | ✅ | ✅ | ✅ | ✅ | ✅ |
| exclusiveMinimum | - | - | ✅ | ✅ | ✅ | ✅ | ✅ |
| if-then-else | - | - | - | ✅ | ✅ | ✅ | ✅ |
| infinite-loop-detection | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| items | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| maxContains | - | - | - | - | ✅ | ✅ | ✅ |
| maximum | 13/14 | 13/14 | ✅ | ✅ | ✅ | ✅ | ✅ |
| maxItems | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| maxLength | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| maxProperties | - | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| minContains | - | - | - | - | ✅ | ✅ | ✅ |
| minimum | 12/13 | 16/17 | ✅ | ✅ | ✅ | ✅ | ✅ |
| minItems | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| minLength | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| minProperties | - | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| multipleOf | - | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| not | - | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| oneOf | - | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| pattern | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| patternProperties | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| prefixItems | - | - | - | - | - | ✅ | ✅ |
| properties | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| propertyNames | - | - | ✅ | ✅ | ✅ | ✅ | ✅ |
| recursiveRef | - | - | - | - | ✅ | - | - |
| ref | 23/27 | 37/45 | 67/70 | 75/78 | 79/81 | 77/79 | 77/79 |
| required | 3/4 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| type | 73/80 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| unevaluatedItems | - | - | - | - | ✅ | 65/71 | 64/71 |
| unevaluatedProperties | - | - | - | - | ✅ | ✅ | 124/125 |
| uniqueItems | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
The following table shows compile performance for various JSON Schema structures using AJV8 as a basis for comparison.
┌──────────────────────┬─────────────┬─────────────┐
│ Compile │ TB1X │ AJV8 │
├──────────────────────┼─────────────┼─────────────┤
│ Boolean │ 29.2K ops/s │ 7.1K ops/s │
│ Number │ 34.5K ops/s │ 7.6K ops/s │
│ String │ 48.9K ops/s │ 8.7K ops/s │
│ Null │ 39.6K ops/s │ 7.6K ops/s │
│ Literal_String │ 46.8K ops/s │ 6.8K ops/s │
│ Literal_Number │ 48.3K ops/s │ 7.4K ops/s │
│ Literal_Boolean │ 48.8K ops/s │ 7.3K ops/s │
│ Pattern │ 32.5K ops/s │ 6.1K ops/s │
│ Object_Open │ 6.6K ops/s │ 1.4K ops/s │
│ Object_Close │ 7.6K ops/s │ 1K ops/s │
│ Object_Vector3 │ 20.8K ops/s │ 2.8K ops/s │
│ Object_Basis3 │ 7.5K ops/s │ 1K ops/s │
│ Intersect_And │ 23K ops/s │ 3.9K ops/s │
│ Intersect_Structural │ 8.7K ops/s │ 1.2K ops/s │
│ Union_Or │ 17.9K ops/s │ 3.4K ops/s │
│ Union_Structural │ 11.3K ops/s │ 2.1K ops/s │
│ Tuple_Values │ 9.6K ops/s │ 2.1K ops/s │
│ Tuple_Objects │ 2.1K ops/s │ 350 ops/s │
│ Array_Numbers_4 │ 33.6K ops/s │ 4.2K ops/s │
│ Array_Numbers_8 │ 39K ops/s │ 3.7K ops/s │
│ Array_Numbers_16 │ 29.6K ops/s │ 3.8K ops/s │
│ Array_Objects_Open │ 7.7K ops/s │ 833 ops/s │
│ Array_Objects_Close │ 7.6K ops/s │ 860 ops/s │
└──────────────────────┴─────────────┴─────────────┘The following tables shows validation performance for various JSON Schema structures using AJV8 as a basis for comparison.
┌──────────────────────┬──────────────┬──────────────┐
│ Validate │ TB1X │ AJV8 │
├──────────────────────┼──────────────┼──────────────┤
│ Boolean │ 192.2M ops/s │ 189.5M ops/s │
│ Number │ 112.4M ops/s │ 61M ops/s │
│ String │ 113.7M ops/s │ 64.1M ops/s │
│ Null │ 112.8M ops/s │ 64.9M ops/s │
│ Literal_String │ 108M ops/s │ 62.9M ops/s │
│ Literal_Number │ 113.5M ops/s │ 63.2M ops/s │
│ Literal_Boolean │ 109.2M ops/s │ 64.1M ops/s │
│ Pattern │ 26.5M ops/s │ 22.4M ops/s │
│ Object_Open │ 78M ops/s │ 47.2M ops/s │
│ Object_Close │ 38.6M ops/s │ 27.6M ops/s │
│ Object_Vector3 │ 91M ops/s │ 51.3M ops/s │
│ Object_Basis3 │ 41.1M ops/s │ 27.4M ops/s │
│ Intersect_And │ 107.6M ops/s │ 59.9M ops/s │
│ Intersect_Structural │ 83.6M ops/s │ 46.3M ops/s │
│ Union_Or │ 95M ops/s │ 7.9M ops/s │
│ Union_Structural │ 84.5M ops/s │ 52.3M ops/s │
│ Tuple_Values │ 74.7M ops/s │ 53M ops/s │
│ Tuple_Objects │ 32.9M ops/s │ 22.3M ops/s │
│ Array_Numbers_4 │ 93.3M ops/s │ 55.1M ops/s │
│ Array_Numbers_8 │ 90.3M ops/s │ 50.8M ops/s │
│ Array_Numbers_16 │ 76.8M ops/s │ 39.6M ops/s │
│ Array_Objects_Open │ 28.7M ops/s │ 20.4M ops/s │
│ Array_Objects_Close │ 10.3M ops/s │ 10.8M ops/s │
└──────────────────────┴──────────────┴──────────────┘TypeBox ships two distinct versions that span two generations of the TypeScript compiler.
| TypeBox | TypeScript | Description |
|---|---|---|
| 1.x | 6.0 - 7.0+ | Latest. Developed against the TypeScript 7 native compiler. Provides advanced type inference and native JSON Schema 2020-12 support. Includes backwards compatibility with 0.x types. ESM only. |
| 0.x | 5.0 - 6.0 | LTS. Developed against older TypeScript versions and actively maintained under Long Term Support. Compatible with both ESM and CJS. Issues should be submitted to the Sinclair TypeBox repository. |
TypeBox is open to community contribution. Please ensure you submit an issue before submitting a pull request. The TypeBox project prefers open community discussion before accepting new features.
