Runtypes allow you to take values about which you have no assurances and check that they conform to some type A. This is done by means of composable type validators of primitives, literals, arrays, tuples, objects, unions, intersections and more.
npm install --save runtypesSuppose you have objects which represent asteroids, planets, ships and crew members. In TypeScript, you might write their types like so:
type Vector = [number, number, number]
type Asteroid = {
type: "asteroid"
location: Vector
mass: number
}
type Planet = {
type: "planet"
location: Vector
mass: number
population: number
habitable: boolean
}
type Rank = "captain" | "first mate" | "officer" | "ensign"
type CrewMember = {
name: string
age: number
rank: Rank
home: Planet
}
type Ship = {
type: "ship"
location: Vector
mass: number
name: string
crew: CrewMember[]
}
type SpaceObject = Asteroid | Planet | ShipIf the objects which are supposed to have these shapes are loaded from some external source, perhaps a JSON file, we need to validate that the objects conform to their specifications. We do so by building corresponding Runtypes in a very straightforward manner:
import { Boolean, Number, String, Literal, Array, Tuple, Object, Union } from "runtypes"
const Vector = Tuple(Number, Number, Number)
const Asteroid = Object({
type: Literal("asteroid"),
location: Vector,
mass: Number,
})
const Planet = Object({
type: Literal("planet"),
location: Vector,
mass: Number,
population: Number,
habitable: Boolean,
})
const Rank = Union(Literal("captain"), Literal("first mate"), Literal("officer"), Literal("ensign"))
const CrewMember = Object({
name: String,
age: Number,
rank: Rank,
home: Planet,
})
const Ship = Object({
type: Literal("ship"),
location: Vector,
mass: Number,
name: String,
crew: Array(CrewMember),
})
const SpaceObject = Union(Asteroid, Planet, Ship)(See the examples directory for an expanded version of this.)
Now if we are given a putative SpaceObject we can validate it like so:
// spaceObject: SpaceObject
const spaceObject = SpaceObject.check(value)If the object doesn't conform to the type specification, check will throw an exception.
When it fails to validate, your runtype emits a ValidationError object that contains detailed information that describes what's the problem. Following properties are available in the object:
name: Always"ValidationError"message: Astringthat summarizes the problem overallcode: AFailcodethat categorizes the problemdetails: An object that describes which property was invalid precisely; only for complex runtypes (e.g.Object,Array, and the like)
If you want to inform your users about the validation error, it's strongly discouraged to rely on the format of message property in your code, as it may change across minor versions for readability thoughts. Instead of parsing message, you should use code and/or details property to programmatically inspect the validation error, and handle other stuff such as i18n.
The inferred type of Asteroid in the above example is a subtype of
Runtype.Common<{
type: "asteroid"
location: [number, number, number]
mass: number
}>That is, it's a Runtype.Common<Asteroid>, and you could annotate it as such. But we don't really have to define the Asteroid type at all now, because the inferred type is correct. Defining each of your types twice, once at the type level and then again at the value level, is a pain and not very DRY. Fortunately you can define a static Asteroid type which is an alias to the Runtype-derived type like so:
type Asteroid = Static<typeof Asteroid>which achieves the same result as
type Asteroid = {
type: "asteroid"
location: [number, number, number]
mass: number
}Instead of getting it to be inferred, you should be able to create a runtype that corresponds to a static type predefined somewhere. In such case you can statically ensure that your runtype conforms to the specification, by using .conform<T>():
type Specification = {
foo: string
bar?: string
}
const Correct = Object({
foo: String,
bar: String.optional(),
}).conform<Specification>()
// @ts-expect-error: should fail
const Wrong = Object({
foo: String,
bar: String,
}).conform<Specification>()The error message on the wrong definition might be verbose like below, but you'll eventually find it contains where is the wrong piece if you scroll down the wall of text.
The 'this' context of type 'WithUtilities<{ foo: String; bar: String; }>' is not assignable to method's 'this' of type 'Conform<Specification>'.
Type 'WithUtilities<{ foo: String; bar: String; }>' is not assignable to type 'Conformance<Specification>'.
Types of property '[RuntypeConformance]' are incompatible.
Type '(StaticTypeOfThis: { foo: string; bar: string; }) => { foo: string; bar: string; }' is not assignable to type '(StaticTypeOfThis: Specification) => Specification'.
Types of parameters 'StaticTypeOfThis' and 'StaticTypeOfThis' are incompatible.
Type 'Specification' is not assignable to type '{ foo: string; bar: string; }'.
Property 'bar' is optional in type 'Specification' but required in type '{ foo: string; bar: string; }'.
Runtypes provide a guard function as the guard method:
const disembark = (value: unknown) => {
if (SpaceObject.guard(value)) {
// value: SpaceObject
if (value.type === "ship") {
// value: Ship
value.crew = []
}
}
}Runtypes provide an assertion function as the assert method:
const disembark = (value: unknown) => {
try {
SpaceObject.assert(value)
// value: SpaceObject
if (value.type === "ship") {
// value: Ship
value.crew = []
}
} catch (error) {}
}This might be uncomfortable that TypeScript requires you to manually write the type annotation for your runtype.
The Union runtype offers the ability to do type-safe, exhaustive case analysis across its variants using the match method:
const isHabitable = SpaceObject.match(
asteroid => false,
planet => planet.habitable,
ship => true,
)
if (isHabitable(spaceObject)) {
// ...
}There's also a top-level match function which allows testing an ad-hoc sequence of runtypes. You should use it along with when helper function to enable type inference of the parameters of the case functions:
const makeANumber = match(
when(Number, n => n * 3),
when(Boolean, b => (b ? 1 : 0)),
when(String, s => s.length),
)
makeANumber(9) // = 27To allow the function to be applied to anything and then handle match failures, simply use an Unknown case at the end:
const makeANumber = match(
when(Number, n => n * 3),
when(Boolean, b => (b ? 1 : 0)),
when(String, s => s.length),
when(Unknown, () => 42),
)Beyond mere type checking, we can add arbitrary runtime constraints to a Runtype:
const PositiveNumber = Number.withConstraint(n => n > 0)
PositiveNumber.check(-3) // Throws error: Failed constraint checkYou can provide more descriptive error messages for failed constraints by returning a string instead of false:
const PositiveNumber = Number.withConstraint(n => n > 0 || `${n} is not positive`)
PositiveNumber.check(-3) // Throws error: -3 is not positiveConstraint checking narrows down the original type to a subtype of it. This should be reflected on the static type. You can pass the desired type as the type argument:
const TheAnswer = Literal(42)
const WithConstraint = Number.withConstraint<42>(TheAnswer.guard)
type WithConstraint = Static<typeof WithConstraint> // 42Alternatively, you can directly wire up the TypeScript's own facility to narrow down types: guard functions and assertion functions. There're corresponding methods on a runtype, so choose the most concise one:
const WithGuard = Number.withGuard(TheAnswer.guard)
type WithGuard = Static<typeof WithGuard> // 42
const WithAssertion = Number.withAssertion(TheAnswer.assert)
type WithAssertion = Static<typeof WithAssertion> // 42If you want to provide custom error messages while narrowing static types, you can throw string or Error from a constraint, guard, or assertion function. Actually, returning a string from a function passed to withConstraint is supported by this exception handling internally.
Too often there might be cases you can't express desired types exactly in TypeScript, such as the type for positive numbers. In such cases you should at least express them as branded types.
const PositiveNumber = Number.withConstraint(n => n > 0).withBrand("PositiveNumber")withBrand modifier is also useful when you want to give your runtype a custom name, which will be used in error messages.
The Template runtype validates that a value is a string that conforms to the template.
You can use the familiar syntax to create a Template runtype:
const T = Template`foo${Literal("bar")}baz`But then the type inference won't work:
type T = Static<typeof T> // stringBecause TS doesn't provide the exact string literal type information (["foo", "baz"] in this case) to the underlying function. See the issue microsoft/TypeScript#33304, especially this comment microsoft/TypeScript#33304 (comment) we hope to be implemented.
If you want the type inference rather than the tagged syntax, you have to manually write a function call:
const T = Template(["foo", "baz"] as const, Literal("bar"))
type T = Static<typeof T> // "foobarbaz"As a convenient solution for this, it also supports another style of passing arguments:
const T = Template("foo", Literal("bar"), "baz")
type T = Static<typeof T> // "foobarbaz"You can pass various things to the Template constructor, as long as they are assignable to string | number | bigint | boolean | null | undefined and the corresponding Runtypes:
// Equivalent runtypes
Template(Literal("42"))
Template(42)
Template(Template("42"))
Template(4, "2")
Template(Literal(4), "2")
Template(String.withConstraint(s => s === "42"))
Template(
Intersect(
Number.withConstraint(n => n === 42),
String.withConstraint(s => s.length === 2),
// `Number`s in `Template` accept alternative representations like `"0x2A"`,
// thus we have to constraint the length of string, to accept only `"42"`
),
)Trivial items such as bare literals, Literals, and single-element Unions and Intersects are all coerced into strings at the creation time of the runtype. Additionally, Unions of such runtypes are converted into RegExp patterns like (?:foo|bar|...), so we can assume Union of Literals is a fully supported runtype in Template.
A Template internally constructs a RegExp to parse strings. This can lead to a problem if it contains multiple non-literal runtypes:
const UpperCaseString = String.withConstraint(s => s === s.toUpperCase(), {
name: "UpperCaseString",
})
const LowerCaseString = String.withConstraint(s => s === s.toLowerCase(), {
name: "LowerCaseString",
})
Template(UpperCaseString, LowerCaseString) // DON'T DO THIS!The only thing we can do for parsing such strings correctly is brute-forcing every single possible combination until it fulfills all the constraints, which must be hardly done. Actually Template treats String runtypes as the simplest RegExp pattern .* and the “greedy” strategy is always used, that is, the above runtype won't work expectedly because the entire pattern is just ^(.*)(.*)$ and the first .* always wins. You have to avoid using Constraint this way, and instead manually parse it using a single Constraint which covers the entire string.
You can spread a Tuple or an Array within arguments of Tuple.
const T = Tuple(Literal(0), ...Tuple(Literal(1), Literal(2)), Literal(3))
type T = Static<typeof T> // [0, 1, 2, 3]
const U = Tuple(Literal(0), ...Array(Literal(1)), Literal(2))
type U = Static<typeof U> // [0, ...1[], 2]If you have access to the class that you want to test values with the instanceof operator, then the InstanceOf runtype is exactly what you're looking for. Usage is straightforward:
class ObjectId { ... };
const ObjectIdChecker = InstanceOf(ObjectId);
ObjectIdChecker.check(value);Branded types is a way to emphasize the uniqueness of a type. This is useful until we have nominal types:
const Username = String.withBrand("Username")
const Password = String.withBrand("Password").withConstraint(
str => str.length >= 8 || "Too short password",
)
const signIn = Contract({
receives: Tuple(Username, Password),
returns: Unknown,
}).enforce((username, password) => {
/*...*/
})
const username = Username.check("[email protected]")
const password = Password.check("12345678")
// Static type OK, runtime OK
signIn(username, password)
// Static type ERROR, runtime OK
signIn(password, username)
// Static type ERROR, runtime OK
signIn("[email protected]", "12345678")Object runtypes should be able to express optional properties. There's a modifier to do that: .optional().
Object({ x: Number.optional() })You must be aware of the difference between Object({ x: Union(String, Undefined) }) and Object({ x: String.optional() }); the former means “x must be present, and must be string or undefined”, while the latter means “x can be present or absent, but must be string if present”.
It's strongly discouraged to disable "exactOptionalPropertyTypes" in the tsconfig; if you do so, the correspondence between runtypes and the inferred static types get lost. We can't respect tsconfig at runtime, so runtypes always conform the behavior "exactOptionalPropertyTypes": true, in favor of the expressiveness.
Object has a modifier to perform exact object validation: .exact().
const O = Object({ x: Number }).exact()
O.guard({ x: 42 }) // true
O.guard({ x: 42, y: 24 }) // falseNote that TypeScript doesn't have exact types at the moment, so it's recommended to wrap your exact Object runtype within a Brand to at least prevent the unexpected behavior of the inferred static type:
const x0 = { x: 42 }
const x1 = { x: 42, y: 24 }
const O = Object({ x: Number }).exact()
type O = Static<typeof O>
const o0: O = x0
const o1: O = x1 // You would not want this to be possible.
globalThis.Object.hasOwn(o1, "y") === true
const P = O.withBrand("P")
type P = Static<typeof P>
const p0: P = P.check(x0) // Branded types require explicit assertion.
const p1: P = P.check(x1) // So this won't accidentally pass at runtime.You should beware that Object validation only respects enumerable own keys; thus if you want to completely eliminate extra properties that may be non-enumerable or inherited, use parse method.
Every runtype has the withParser and parse methods that offer the functionality to transform validated values automatically.
const O = Object({ x: String.withParser(parseInt).default(42) })
type OStatic = Static<typeof O> // { x: string }
type OParsed = Parsed<typeof O> // { x: number }
O.parse({ x: "42" }).x === 42The .default(...) modifier works the same as .optional() for mere validation, but for parsing, it works as falling back to the value if the property was absent.
O.parse({}).x === 42Extraneous properties are not copied to the resulting value.
"y" in O.parse({ y: "extra" }) === falseWhile parse returns a new value, traditional validation methods such as check don't change their semantics even with parsers.
const o: OStatic = { x: "42" }
o === O.check(o)In an Object, an Array, and a Tuple, Parsers will work just as you'd expect.
In a Template, parsing can work like this:
const TrueToFalse = Literal("true").withParser(() => "false" as const)
const Value = Template("value: ", TrueToFalse)
Value.parse("value: true") === "value: false"In a Union, the first succeeding runtype returns a value and further alternatives are not executed at all:
const Flip = Union(
Boolean.withParser(b => !b),
Boolean.withParser(b => !!b),
)
Flip.parse(true) === falseIn an Intersect, the last runtype returns a value and preceding intersectees are executed but results are just discarded:
const FlipFlip = Intersect(
Boolean.withParser(b => !b),
Boolean.withParser(b => !!b),
)
FlipFlip.parse(true) === trueArray and Object runtypes have a special function .asReadonly(), that returns the same runtype but the static counterpart is readonly.
For example:
const Asteroid = Object({
type: Literal("asteroid"),
location: Vector,
mass: Number,
}).asReadonly()
type Asteroid = Static<typeof Asteroid>
// { readonly type: 'asteroid', readonly location: Vector, readonly mass: number }
const AsteroidArray = Array(Asteroid).asReadonly()
type AsteroidArray = Static<typeof AsteroidArray>
// readonly Asteroid[]Object runtype has the methods .pick() and .omit(), which will return a new Object with or without specified fields (see Example section for detailed definition of Rank and Planet):
const CrewMember = Object({
name: String,
age: Number,
rank: Rank,
home: Planet,
})
const Visitor = CrewMember.pick("name", "home")
type Visitor = Static<typeof Visitor> // { name: string; home: Planet; }
const Background = CrewMember.omit("name")
type Background = Static<typeof Background> // { age: number; rank: Rank; home: Planet; }Also you can use .extend() to get a new Object with extended fields:
const PetMember = CrewMember.extend({
species: String,
})
type PetMember = Static<typeof PetMember>
// { name: string; age: number; rank: Rank; home: Planet; species: string; }It is capable of reporting compile-time errors if any field is not assignable to the base runtype. You can suppress this error by using @ts-ignore directive or .omit() before, and then you'll get an incompatible version from the base Object.
const WrongMember = CrewMember.extend({
rank: Literal("wrong"),
// Type '"wrong"' is not assignable to type '"captain" | "first mate" | "officer" | "ensign"'.
})You may want to provide additional properties along with your runtype, such as the default value and utility functions. This can be easily achieved by the with method.
const Seconds = Number.withBrand("Seconds").with({
toMilliseconds: (seconds: Seconds) => (seconds * 1000) as Milliseconds,
})
type Seconds = Static<typeof Seconds>
const Milliseconds = Number.withBrand("Milliseconds").with({
toSeconds: (milliseconds: Milliseconds) => (milliseconds / 1000) as Seconds,
})
type Milliseconds = Static<typeof Milliseconds>Sometimes defining additional properties requires access to the original runtype itself statically or dynamically:
// Bummer, this won't work because of the circular reference.
const pH = Number.withBrand("pH").with({ default: 7 as pH })
type pH = Static<typeof pH>In such cases, you have to receive the original runtype by passing a function instead:
const pH = Number.withBrand("pH").with(self => ({
default: 7 as Static<typeof self>,
}))
type pH = Static<typeof pH>Runtypes along with constraint checking are a natural fit for enforcing function contracts. You can construct a contract from runtypes for the parameters and return type of the function:
const divide = Contract({
// This must be a runtype for arrays, just like annotating a rest parameter in TS.
receives: Tuple(
Number,
Number.withConstraint(n => n !== 0 || "division by zero"),
),
returns: Number,
}).enforce((n, m) => n / m)
divide(10, 2) // 5
divide(10, 0) // Throws error: division by zeroContracts can work with Parser runtypes:
const ParseInt = String.withParser(parseInt)
const contractedFunction = Contract({
receives: Array(ParseInt),
returns: Array(ParseInt),
}).enforce((...args) => args.map(globalThis.String))
contractedFunction("42", "24") // [42, 24]- generate-runtypes Generates runtypes from structured data. Useful for code generators
- json-to-runtypes Generates runtypes by parsing example JSON data
- rest.ts Allows building type safe and runtime-checked APIs
- runtypes-generate Generates random data by
Runtypefor property-based testing - runtyping Generate runtypes from static types & JSON schema
- schemart Generate runtypes from your database schema.
