Generic type variance in Typescript

Introduction

Generic type variance helps control how types relate when used with generics. It makes your code safer and more flexible by defining if types can be replaced by their subtypes or supertypes.

When you want to allow a function to accept a generic type or its subtypes safely.

When you design reusable components that work with different but related types.

When you want to prevent errors by restricting how generic types can be substituted.

When working with collections like arrays or lists that hold generic types.

When defining APIs that should accept or return types in a controlled way.

Syntax

Typescript

// TypeScript does not support 'in' or 'out' keywords like some languages
// Covariance is expressed with readonly properties
// Contravariance is expressed with function parameter types

TypeScript does not use explicit 'in' or 'out' keywords like some languages.

Covariance often happens with readonly properties, contravariance with function parameters.

Examples

This shows covariance: you can use a subtype where a supertype is expected.

Typescript

interface ReadonlyArray<T> {
  readonly [index: number]: T;
}

Function parameters are contravariant, meaning you can use a supertype where a subtype is expected.

Typescript

type Handler<T> = (arg: T) => void;
// Function parameters are contravariant in TypeScript

Covariance allows passing ReadonlyArray<Dog> where ReadonlyArray<Animal> is expected.

Typescript

function processAnimals(animals: ReadonlyArray<Animal>) {}

const dogs: ReadonlyArray<Dog> = [{ name: 'Buddy' }];
processAnimals(dogs);

Sample Program

This example shows covariance: a ReadonlyArray<Dog> can be passed to a function expecting ReadonlyArray<Animal> safely.

Typescript

class Animal {
  name: string;
  constructor(name: string) {
    this.name = name;
  }
}

class Dog extends Animal {
  bark() {
    console.log(`${this.name} says Woof!`);
  }
}

function feedAnimals(animals: ReadonlyArray<Animal>) {
  animals.forEach(animal => console.log(`Feeding ${animal.name}`));
}

const dogs: ReadonlyArray<Dog> = [new Dog('Buddy'), new Dog('Max')];
feedAnimals(dogs);

Important Notes

Covariance means you can use a more specific type where a more general type is expected.

Contravariance means you can use a more general type where a more specific type is expected.

TypeScript infers variance mostly from how types are used (e.g., readonly properties for covariance).

Summary

Generic type variance controls how generic types relate to each other.

Covariance allows substituting a subtype for a supertype safely.

Contravariance allows substituting a supertype for a subtype safely.

Practice

1.

Which of the following best describes covariance in TypeScript generic types?

easy

A. It allows a generic type to accept a supertype where a subtype is expected.

B. It allows a generic type to accept a subtype where a supertype is expected.

C. It restricts generic types to only accept exact types without substitution.

D. It means generic types cannot be assigned to each other regardless of relation.

Solution

1. Step 1: Understand covariance meaning

   Covariance means you can use a more specific type (subtype) where a more general type (supertype) is expected.

2. Step 2: Apply to generic types

   In TypeScript, covariance allows substituting a subtype for a supertype safely in generics.

3. Final Answer:

   It allows a generic type to accept a subtype where a supertype is expected. -> Option B

4. Quick Check:

   Covariance = subtype for supertype [OK]

Hint: Covariance means subtype can replace supertype [OK]

Common Mistakes:

• Confusing covariance with contravariance
• Thinking covariance restricts substitution
• Assuming covariance means exact type only

2.

Which of the following is the correct syntax to declare a generic function with contravariant parameter in TypeScript?

function process<T>(input: T): void { /* ... */ }

easy

A. function process(input: T): void { }

B. function process(input: T): void { }

C. function process<T>(input: T): void { }

D. function process(input: T): void { }

Solution

1. Step 1: Recall TypeScript generic variance syntax

   TypeScript does not use in or out keywords like some languages for variance.

2. Step 2: Check valid generic function syntax

   The correct syntax is simply function process<T>(input: T): void { } without variance keywords.

3. Final Answer:

   function process<T>(input: T): void { } -> Option C

4. Quick Check:

   No variance keywords in TypeScript generics [OK]

Hint: TypeScript uses no in/out keywords for generics [OK]

Common Mistakes:

• Using in/out keywords from other languages
• Trying to extend non-existent variance types
• Assuming variance is declared in generic syntax

3.

Consider the following TypeScript code:

interface Animal { name: string; }
interface Dog extends Animal { bark(): void; }

function feedAnimals(animals: ReadonlyArray<Animal>) { /* ... */ }

const dogs: Dog[] = [{ name: 'Rex', bark: () => console.log('Woof!') }];
feedAnimals(dogs);

What is the output or result of calling feedAnimals(dogs)?

medium

A. It causes a compile-time error due to type incompatibility.

B. It compiles but silently ignores the dogs array.

C. It runs but throws a runtime error because of type mismatch.

D. It compiles and runs without error because ReadonlyArray is covariant.

Solution

1. Step 1: Understand ReadonlyArray variance

   ReadonlyArray in TypeScript is covariant, so Dog[] can be used where ReadonlyArray<Animal> is expected.

2. Step 2: Check function call validity

   Passing dogs (Dog[]) to feedAnimals(animals: ReadonlyArray<Animal>) is allowed and compiles without error.

3. Final Answer:

   It compiles and runs without error because ReadonlyArray is covariant. -> Option D

4. Quick Check:

   ReadonlyArray is covariant = safe assignment [OK]

Hint: ReadonlyArray is covariant, so subtype arrays pass [OK]

Common Mistakes:

• Assuming arrays are invariant by default
• Confusing mutable arrays with ReadonlyArray
• Expecting runtime errors from type mismatch

4.

Identify the error in the following TypeScript code related to generic variance:

interface Printer<T> {
  print(value: T): void;
}

const animalPrinter: Printer<Animal> = {
  print: (animal: Animal) => console.log(animal.name)
};

const dogPrinter: Printer<Dog> = animalPrinter;
dogPrinter.print({ name: 'Buddy', bark: () => console.log('Woof!') });

medium

A. Assigning animalPrinter to dogPrinter causes a type error due to invariance.

B. The print method signature is incorrect and causes a syntax error.

C. There is no error; the code runs and prints 'Buddy'.

D. The dogPrinter variable is not declared properly.

Solution

1. Step 1: Analyze variance of Printer<T>

   Printer<T> has a method accepting T as parameter, making it contravariant in T.

2. Step 2: Check assignment direction

   Assigning Printer<Animal> to Printer<Dog> is invalid because contravariance requires the opposite direction.

3. Final Answer:

   Assigning animalPrinter to dogPrinter causes a type error due to invariance. -> Option A

4. Quick Check:

   Function parameter generics are contravariant [OK]

Hint: Function parameters are contravariant; assignment direction matters [OK]

Common Mistakes:

• Ignoring contravariance in function parameters
• Assuming generic interfaces are always covariant
• Confusing parameter and return type variance

5.

You want to create a generic event handler interface in TypeScript that accepts events of type T. The handler should be able to accept handlers for supertypes of T (contravariance). Which of the following interface declarations correctly models this behavior?

hard

A. type EventHandler<T> = (event: T) => void;

B. interface EventHandler<in T> { handle(event: T): void; }

C. interface EventHandler<T> { handle(event: T): T; }

D. interface EventHandler<T> { handle(event: T): void; }

Solution

1. Step 1: Understand contravariance in function parameters

   Function parameters are contravariant, so a function type like (event: T) => void models contravariance naturally.

2. Step 2: Compare interface and type alias

   Interfaces with methods are invariant by default in TypeScript, but function types are contravariant in parameters, so using a type alias for a function is correct.

3. Step 3: Check invalid syntax options

   interface EventHandler<in T> { handle(event: T): void; } uses in keyword which TypeScript does not support; interface EventHandler<T> { handle(event: T): T; } changes return type, losing contravariance.

4. Final Answer:

   type EventHandler<T> = (event: T) => void; -> Option A

5. Quick Check:

   Function types are contravariant in parameters [OK]

Hint: Use function types for contravariant parameters [OK]

Common Mistakes:

• Using unsupported 'in' keyword in generics
• Assuming interfaces are contravariant by default
• Changing return type affects variance incorrectly