loading...
Cover image for Advanced TypeScript Types cheat sheet (with examples)

Advanced TypeScript Types cheat sheet (with examples)

ibrahima92 profile image Ibrahima Ndaw Originally published at ibrahima-ndaw.com ・8 min read

TypeScript is a typed language that allows you to specify the type of variables, function parameters, returned values, and object properties.

Here an advanced TypeScript Types cheat sheet with examples.

Let's dive in

Originally posted on my blog

Intersection Types

An intersection type is a way of combining multiple types into one. Meaning that you can merge a given type A with a type B or more and get back a single type with all properties.

type LeftType = {
  id: number
  left: string
}

type RightType = {
  id: number
  right: string
}

type IntersectionType = LeftType & RightType

function showType(args: IntersectionType) {
  console.log(args)
}

showType({ id: 1, left: "test", right: "test" })
// Output: {id: 1, left: "test", right: "test"}

As you can see, IntersectionType combines two types - LeftType and RightType and use the & sign to construct the intersection type.

Union Types

Union types allow you to have different types annotation within a given variable.

type UnionType = string | number

function showType(arg: UnionType) {
  console.log(arg)
}

showType("test")
// Output: test

showType(7)
// Output: 7

The function showType is a union type that accepts both strings and numbers as a parameter.

Generic Types

A generic type is a way of reusing part of a given type. It helps to capture the type T passed in as a parameter.

function showType<T>(args: T) {
  console.log(args)
}

showType("test")
// Output: "test"

showType(1)
// Output: 1

To construct a generic type, you need to use the brackets and pass T as a parameter.
Here, I use T (the name is up to you) and then, call the function showType twice with different type annotations because it's generic - it can be reused.

interface GenericType<T> {
  id: number
  name: T
}

function showType(args: GenericType<string>) {
  console.log(args)
}

showType({ id: 1, name: "test" })
// Output: {id: 1, name: "test"}

function showTypeTwo(args: GenericType<number>) {
  console.log(args)
}

showTypeTwo({ id: 1, name: 4 })
// Output: {id: 1, name: 4}

Here, we have another example that has an interface GenericType which receives a generic type T. And since it's reusable, we can call it with first a string, and then a number.

interface GenericType<T, U> {
  id: T
  name: U
}

function showType(args: GenericType<number, string>) {
  console.log(args)
}

showType({ id: 1, name: "test" })
// Output: {id: 1, name: "test"}

function showTypeTwo(args: GenericType<string, string[]>) {
  console.log(args)
}

showTypeTwo({ id: "001", name: ["This", "is", "a", "Test"] })
// Output: {id: "001", name: Array["This", "is", "a", "Test"]}

A generic type can receive several arguments. Here, we pass in two parameters: T and U, and then use them as type annotations for the properties. That said, we can now use the interface and provide different types as argument.

Utility Types

TypeScript provides handy built-in utilities that help to manipulate types easily. To use them, you need to pass into the <> the type you want to transform.

Partial

  • Partial<T>

Partial allows you to make all properties of the type T optional. It will add a ? mark next to every field.

interface PartialType {
  id: number
  firstName: string
  lastName: string
}

function showType(args: Partial<PartialType>) {
  console.log(args)
}

showType({ id: 1 })
// Output: {id: 1}

showType({ firstName: "John", lastName: "Doe" })
// Output: {firstName: "John", lastName: "Doe"}

As you can see, we have an interface PartialType which is used as type annotation for the parameters received by the function showType(). And to make the properties optional, we have to use the Partial keyword and pass in the type PartialType as an argument. That said, now all fields become optional.

Required

  • Required<T>

Unlike Partial, the Required utility makes all properties of the type T required.

interface RequiredType {
  id: number
  firstName?: string
  lastName?: string
}

function showType(args: Required<RequiredType>) {
  console.log(args)
}

showType({ id: 1, firstName: "John", lastName: "Doe" })
// Output: { id: 1, firstName: "John", lastName: "Doe" }

showType({ id: 1 })
// Error: Type '{ id: number: }' is missing the following properties from type 'Required<RequiredType>': firstName, lastName

The Required utility will make all properties required even if we make them optional first before using the utility. And if a property is omitted, TypeScript will throw an error.

Readonly

  • Readonly<T>

This utility type will transform all properties of the type T in order to make them not reassignable with a new value.

interface ReadonlyType {
  id: number
  name: string
}

function showType(args: Readonly<ReadonlyType>) {
  args.id = 4
  console.log(args)
}

showType({ id: 1, name: "Doe" })
// Error: Cannot assign to 'id' because it is a read-only property.

Here, we use the utility Readonly to make the properties of ReadonlyType not reassignable. That said, if you try to give a new value to one of these fields, an error will be thrown.

Besides that, you can also use the keyword readonly in front of a property to make it not reassignable.

interface ReadonlyType {
  readonly id: number
  name: string
}

Pick

  • Pick<T, K>

It allows you to create a new type from an existing model T by selecting some properties K of that type.

interface PickType {
  id: number
  firstName: string
  lastName: string
}

function showType(args: Pick<PickType, "firstName" | "lastName">) {
  console.log(args)
}

showType({ firstName: "John", lastName: "Doe" })
// Output: {firstName: "John"}

showType({ id: 3 })
// Error: Object literal may only specify known properties, and 'id' does not exist in type 'Pick<PickType, "firstName" | "lastName">'

Pick is a bit different from the previous utilities we have already seen. It expects two parameters - T is the type you want to pick elements from and K which is the property you want to select. You can also pick multiple fields by separating them with a pipe(|) symbol.

Omit

  • Omit<T, K>

The Omit utility is the opposite of the Pick type. And instead of selecting elements, it will remove K properties from the type T.

interface PickType {
  id: number
  firstName: string
  lastName: string
}

function showType(args: Omit<PickType, "firstName" | "lastName">) {
  console.log(args)
}

showType({ id: 7 })
// Output: {id: 7}

showType({ firstName: "John" })
// Error: Object literal may only specify known properties, and 'firstName' does not exist in type 'Pick<PickType, "id">'

This utility is similar to the way Pick works. It expects the type and the properties to omit from that type.

Extract

  • Extract<T, U>

Extract allows you to construct a type by picking properties that are present in two different types. The utility will extract from T all properties that are assignable to U.

interface FirstType {
  id: number
  firstName: string
  lastName: string
}

interface SecondType {
  id: number
  address: string
  city: string
}

type ExtractType = Extract<keyof FirstType, keyof SecondType>
// Output: "id"

Here, we have two types that have in common the property id. And hence by using the Extract keyword, we get back the field id since it's present in both interfaces. And if you have more than one shared field, the utility will extract all similar properties.

Exclude

Unlike Extract, the Exclude utility will construct a type by excluding properties that are already present in two different types. It excludes from T all fields that are assignable to U.

interface FirstType {
  id: number
  firstName: string
  lastName: string
}

interface SecondType {
  id: number
  address: string
  city: string
}

type ExcludeType = Exclude<keyof FirstType, keyof SecondType>

// Output; "firstName" | "lastName"

As you can see here, the properties firstName and lastName are assignable to the SecondType type since they are not present there. And by using the Extract keyword, we get back these fields as expected.

Record

  • Record<K,T>

This utility helps you to construct a type with a set of properties K of a given type T. Record is really handy when it comes to mapping the properties of a type to another one.

interface EmployeeType {
  id: number
  fullname: string
  role: string
}

let employees: Record<number, EmployeeType> = {
  0: { id: 1, fullname: "John Doe", role: "Designer" },
  1: { id: 2, fullname: "Ibrahima Fall", role: "Developer" },
  2: { id: 3, fullname: "Sara Duckson", role: "Developer" },
}

// 0: { id: 1, fullname: "John Doe", role: "Designer" },
// 1: { id: 2, fullname: "Ibrahima Fall", role: "Developer" },
// 2: { id: 3, fullname: "Sara Duckson", role: "Developer" }

The way Record works is relatively simple. Here, it expects a number as a type which is why we have 0, 1, and 2 as keys for the employees variable. And if you try to use a string as a property, an error will be thrown. Next, the set of properties is given by EmployeeType hence the object with the fields id, fullName, and role.

NonNullable

  • NonNullable<T>

It allows you to remove null and undefined from the type T.

type NonNullableType = string | number | null | undefined

function showType(args: NonNullable<NonNullableType>) {
  console.log(args)
}

showType("test")
// Output: "test"

showType(1)
// Output: 1

showType(null)
// Error: Argument of type 'null' is not assignable to parameter of type 'string | number'.

showType(undefined)
// Error: Argument of type 'undefined' is not assignable to parameter of type 'string | number'.

Here, we pass the type NonNullableType as an argument to the NonNullable utility which constructs a new type by excluding null and undefined from that type. That said, if you pass a nullable value, TypeScript will throw an error.

By the way, if you add the --strictNullChecks flag to the tsconfig file, TypeScript will apply non-nullability rules.

Mapped types

Mapped types allow you to take an existing model and transform each of its properties into a new type. Note that some utility types covered earlier are also mapped types.

type StringMap<T> = {
  [P in keyof T]: string
}

function showType(arg: StringMap<{ id: number; name: string }>) {
  console.log(arg)
}

showType({ id: 1, name: "Test" })
// Error: Type 'number' is not assignable to type 'string'.

showType({ id: "testId", name: "This is a Test" })
// Output: {id: "testId", name: "This is a Test"}

StringMap<> will transform whatever types that passed in into a string. That said, if we use it in the function showType(), the parameters received must be a string - otherwise, an error will be thrown by TypeScript.

Type Guards

Type Guards allow you to check the type of a variable or an object with an operator. It's a conditional block that returns a type using typeof, instanceof, or in.

  • typeof
function showType(x: number | string) {
  if (typeof x === "number") {
    return `The result is ${x + x}`
  }
  throw new Error(`This operation can't be done on a ${typeof x}`)
}

showType("I'm not a number")
// Error: This operation can't be done on a string

showType(7)
// Output: The result is 14

As you can see, we have a normal JavaScript conditional block that checks the type of the argument received with typeof. With that in place, you can now guard your type with this condition.

  • instanceof
class Foo {
  bar() {
    return "Hello World"
  }
}

class Bar {
  baz = "123"
}

function showType(arg: Foo | Bar) {
  if (arg instanceof Foo) {
    console.log(arg.bar())
    return arg.bar()
  }

  throw new Error("The type is not supported")
}

showType(new Foo())
// Output: Hello World

showType(new Bar())
// Error: The type is not supported

Like the previous example, this one is also a type guard that checks if the parameter received is part of the Foo class or not and handles it consequently.

  • in
interface FirstType {
  x: number
}
interface SecondType {
  y: string
}

function showType(arg: FirstType | SecondType) {
  if ("x" in arg) {
    console.log(`The property ${arg.x} exists`)
    return `The property ${arg.x} exists`
  }
  throw new Error("This type is not expected")
}

showType({ x: 7 })
// Output: The property 7 exists

showType({ y: "ccc" })
// Error: This type is not expected

The in operator allows you to check whether a property x exists or not on the object received as a parameter.

Conditional Types

It tests two types and selects one of them depending on the outcome of that test.

type NonNullable<T> = T extends null | undefined ? never : T

This example of the NonNullable utility type checks if the type is null or not and handle it depending on that. And as you can note, it uses the JavaScript ternary operator.

Thanks for reading.

You can find other great content like this on my blog or follow me on Twitter to get notified.

Posted on by:

ibrahima92 profile

Ibrahima Ndaw

@ibrahima92

JavaScript enthusiast, Full-stack developer and blogger

Discussion

pic
Editor guide
 

A really well-written article with good examples. Good job!

Would like to add one more thing I like to use and you might find it interesting:

type Maybe<T> = T | null;

function foo(bar: Maybe<string>) {
  if (!bar) {
    // handle if bar is null
  }
  // do stuff if value not null.
}
 

What are some really good real world examples for this? (3,4 examples).

Wouldn't the optional operator be enough?

function foo(bar?: string) {
  if (!bar) {
    // handle if bar is falsy
  }
  // do stuff if value not falsy.
}

The only thing I can think of is when you really need the parameter to be exactly null. But I never have this use case in the projects I work on, so I think something is fishy :))

 

The difference between the two, is that in Maybe the value is required, and in the second one is optional. So, you can have Maybe values and those values would be either defined or not, but would still be required. I have found this to be useful in React.

 

Actually it's just personal preference to have null values instead of undefined.
The example I provided is actually not that great, because I tend to use this more with objects and their properties.

 

You should explicitly check for bar===null as in your case the empty string will also make !bar true

 

you are right, thanks :)

 

Great article! Please remember to use semicolons at the end of the lines.

 

The semicolon is not mandatory, I prefer to not use it. Thanks for reading BTW

 

No semicolon makes it look so much cleaner:)

 

:) so glad I don't have to see or type those anymore

 

Use prettier, you don't even have to type it yourself.

 

It's not mandatory if you wanna go against Google: google.github.io/styleguide/jsguid... . Please reconsider.

Theres a million different style guides, use whatever you want. No one is forced to use a specific one.

 

Congratulations! Really good article, very useful! I would like to add the possibility to create your own type guards on the type guard section, here is an example:

typeguard example

 

Great! A really good example, I will add it to the article.

 

i think : obj is ImportantType is ok, but can be more simpler:

function isImportantType(obj: any): boolean {
  return obj.id && obj.requiredFiled;
}
 

That defeats the purpose of using the type guard altogether. Typescript won’t infer the type from that function. :obj is ImportantType is the what tells typescript what the passed parameters type is. Hence type guard.

thanks for explain

 

But it wouldn't be a type guard though.

thanks for explain

 

Seems typescript is getting overcomplicated.

 

How so? These have been in it for a long time now. You are also not required to use them, since they're mostly syntactic sugar. It does make life easier a lot of times though!

 

Absolutely.

 

Good article, to the point and concise with good examples.

I wish I knew Partial<T> - that I just learnt here before. This is how I implemented it:

export type ObjectWithOptionalProps<T> = { [key in keyof T]?: T[key] };
 

Good article, nice to have everything combined in one place.

 

Thanks for reading too.

 

OMG, what I see! The first well wrote and detailed explanation of the typescript types system!!! Thanks a lot!!!

 

Awesome! I'm glad you find value on it.

 

It is a list of "easy to get" examples of the most interesting TS typing features, it's quite handy having them packed like this, thank you!

 

Great! I'm glad you find value in this article.

 

Really concise summary / recap.

Will share that with my team, thanks a lot !

 

Great! I'm glad this article could help.

 

Thanks for sharing!
I want to translate it into Chinese to help more developers, Can I get your authorization?

I will give credit at the top of the article.

 

Yes, you can translate it. Just make sure to give me credit.