Functype

A Functional Programming Library for TypeScript

Functype is a lightweight functional programming library for TypeScript, drawing inspiration from functional programming paradigms, the Scala Standard Library, and ZIO. It provides a comprehensive set of utilities and abstractions designed to facilitate functional programming within TypeScript applications.

API Documentation · Cheatsheet for Scala / cats / fp-ts users

CLI Documentation

Get LLM-optimized API reference directly in your terminal:

npx functype              # Overview of all types
npx functype Option       # Detailed info for Option type
npx functype interfaces   # All interface definitions
npx functype --json       # JSON output for programmatic use

Core Principles

• Zero Dependencies: No third-party runtime dependencies — just TypeScript
• Immutability: All data structures are immutable, promoting predictable and side-effect-free code
• Type Safety: Leverages TypeScript's type system to ensure compile-time safety
• Composability: Provides abstractions for building complex programs from simple components
• Functional Paradigms: Embraces concepts like monads, functors, and type classes
• Unified Interface: All data structures implement a common hierarchy of interfaces for consistency

Key Features

• Option Type: Handle nullable values with Some and None types
• Either Type: Express computation results with potential failures using Left and Right
• List, Set, Map: Immutable collection types with functional operators
• Try Type: Safely execute operations that might throw exceptions
• Do-notation: Scala-like for-comprehensions with optimized List performance (up to 12x faster than traditional flatMap)
• Task: Handle synchronous and asynchronous operations with error handling
• Http: Typed HTTP fetch wrapper returning IO effects with HttpError ADT, auto content-type detection, and configurable clients
• Lazy: Deferred computation with memoization
• Tuple: Type-safe fixed-length arrays
• Typeable: Runtime type identification with compile-time safety
• Branded Types: Nominal typing in TypeScript's structural type system
• Error Formatting: Utilities for improved error visualization and logging
• Unified Type Classes: Consistent interfaces across all data structures

Installation

# NPM
npm install functype

# Yarn
yarn add functype

# PNPM
pnpm add functype

# Bun
bun add functype

Bundle Size Optimization

Functype is optimized for tree-shaking and offers multiple import strategies to minimize bundle size:

// Selective module imports (recommended for production)
import { Option } from "functype/option"
import { Either } from "functype/either"

// Direct constructor imports (smallest bundle)
import { some, none } from "functype/option"

For detailed optimization strategies, see the Bundle Optimization Guide.

Usage Examples

Option

{@includeCode test/docs/documentation-examples.spec.ts#readme-option-basic}

Either

{@includeCode test/docs/documentation-examples.spec.ts#readme-either-basic}

List

import { List } from "functype"

const numbers = List([1, 2, 3, 4])

// Transform
const doubled = numbers.map((x) => x * 2) // List([2, 4, 6, 8])

// Filter
const evens = numbers.filter((x) => x % 2 === 0) // List([2, 4])

// Reduce
const sum = numbers.foldLeft(0)((acc, x) => acc + x) // 10

// Add/remove elements (immutably)
const withFive = numbers.add(5) // List([1, 2, 3, 4, 5])
const without3 = numbers.remove(3) // List([1, 2, 4])

// Universal container operations
const hasEven = numbers.exists((x) => x % 2 === 0) // true
const firstEven = numbers.find((x) => x % 2 === 0) // Some(2)
const evenCount = numbers.count((x) => x % 2 === 0) // 2

Try

import { Try } from "functype"

// Safely execute code that might throw
const result = Try(() => {
  // Potentially throwing operation
  return JSON.parse('{"name": "John"}')
})

// Handle success/failure
if (result.isSuccess()) {
  console.log("Result:", result.get())
} else {
  console.error("Error:", result.error)
}

// Transform with map (only applies on Success)
const name = result.map((obj) => obj.name)

// Convert to Either
const either = result.toEither()

Lazy

import { Lazy } from "functype"

// Create lazy computations
const expensive = Lazy(() => {
  console.log("Computing...")
  return Math.random() * 1000
})

// Value is computed on first access and memoized
const value1 = expensive.get() // Logs "Computing...", returns number
const value2 = expensive.get() // Returns same number, no log

// Transform lazy values
const doubled = expensive.map((x) => x * 2)
const formatted = doubled.map((x) => `Value: ${x}`)

// Chain computations
const result = Lazy(() => 10)
  .flatMap((x) => Lazy(() => x + 5))
  .map((x) => x * 2)
  .get() // 30

Do-notation (High-Performance For-Comprehensions)

Functype provides generator-based Do-notation for monadic composition, similar to Scala's for-comprehensions, with significant performance advantages for List operations:

import { Do, DoAsync, $ } from "functype"
import { Option, Right, Left, List, Try } from "functype"

// Chain multiple Option operations
const result = Do(function* () {
  const x = yield* $(Option(5)) // Extract value from Option
  const y = yield* $(Option(10)) // Extract value from another Option
  const z = x + y // Regular computation
  return z * 2 // Return final result
})
// result: Option<number> with value 30

// Mix different monad types (with Reshapeable)
const mixed = Do(function* () {
  const a = yield* $(Option(5)) // From Option
  const b = yield* $(Right<string, number>(10)) // From Either
  const c = yield* $(List([15])) // From List
  const d = yield* $(Try(() => 20)) // From Try
  return a + b + c + d
})
// Convert result to desired type
const asOption = mixed.toOption() // Option<number> with value 50

// Error propagation - short-circuits on failure
const validation = Do(function* () {
  const email = yield* $(validateEmail("user@example.com")) // Returns Option
  const user = yield* $(fetchUser(email)) // Returns Either
  const profile = yield* $(loadProfile(user.id)) // Returns Try
  return profile
})
// If any step fails, the entire computation short-circuits

// List comprehensions - up to 12x FASTER than traditional flatMap!
const pairs = Do(function* () {
  const x = yield* $(List([1, 2, 3]))
  const y = yield* $(List([10, 20]))
  return { x, y, product: x * y }
})
// pairs: List with 6 elements (all combinations)

// Performance comparison:
// Traditional: list.flatMap(x => list.flatMap(y => List([{x, y}]))) - slower
// Do-notation: 2.5x to 12x faster for cartesian products!

// Async operations with DoAsync
const asyncResult = await DoAsync(async function* () {
  const user = yield* $(await fetchUserAsync(userId)) // Async Option
  const score = yield* $(await getScoreAsync(user.id)) // Async Either
  const bonus = yield* $(await calculateBonus(score)) // Async Try
  return score + bonus
})

Performance Advantages:

• List Comprehensions: 2.5x to 12x faster than nested flatMap chains
• Optimized for Cartesian Products: Efficient handling of multiple List yields
• Smart Caching: Constructor lookups cached after first type detection
• Inline Helpers: Reduced overhead from repeated type checks

When to Use Do-notation:

✅ Best for:

• Complex List comprehensions (huge performance win!)
• Cartesian products and filtered combinations
• Mixed monad types (leveraging Reshapeable)
• Improved readability for multi-step operations

⚠️ Consider alternatives for:

• Simple 2-3 step Option/Either chains (traditional flatMap is ~2x faster)
• Performance-critical hot paths with simple monads
• Early termination scenarios (flatMap auto-short-circuits more efficiently)

Key Differences from Scala:

• Uses yield* $(monad) instead of x <- monad
• No native guard syntax (use conditions with early return)
• Always returns the type of the first yielded monad
• Mixed types supported via Reshapeable interface

Task

Task v2 provides structured error handling with the Ok/Err pattern, returning TaskOutcome<T> for all operations:

import { Task, Ok, Err, type TaskOutcome } from "functype"

// Task v2: All operations return TaskOutcome<T>
const syncResult = Task().Sync(() => "success")
// Returns: TaskSuccess<string> (extends TaskOutcome<string>)

const asyncResult = await Task().Async(async () => "value")
// Returns: TaskOutcome<string>

// Explicit Ok/Err returns for precise control
const explicitResult = await Task().Async(async (): Promise<TaskOutcome<string>> => {
  if (Math.random() > 0.5) {
    return Ok("success") // Explicit success
  }
  return Err<string>("failed") // Explicit failure
})

// Auto-wrapping: raw values become Ok, thrown errors become Err
const autoWrapped = await Task().Async(async () => {
  if (condition) {
    return "raw value" // Auto-wrapped as Ok("raw value")
  }
  throw new Error("failed") // Auto-wrapped as Err(error)
})

// Error recovery: error handlers can return Ok
const recovered = await Task().Async(
  async () => {
    throw new Error("initial error")
  },
  async (error) => Ok("recovered from error"), // Recovery!
)

// Working with results
if (asyncResult.isSuccess()) {
  console.log(asyncResult.value) // Access the success value
} else {
  console.error(asyncResult.error) // Access the error (Throwable)
}

// Chaining with TaskOutcome
const chainedResult = await Task().Async(async () => {
  const firstResult = await Task().Async(async () => "first")
  if (firstResult.isFailure()) {
    return firstResult // Propagate failure
  }

  const secondResult = await Task().Async(async () => "second")
  if (secondResult.isFailure()) {
    return secondResult
  }

  return Ok(`${firstResult.value} + ${secondResult.value}`)
})

// Converting promise-based functions to Task
const fetchUserAPI = (userId: string): Promise<User> => fetch(`/api/users/${userId}`).then((r) => r.json())

const fetchUser = Task.fromPromise(fetchUserAPI)
// Returns: (userId: string) => Promise<TaskOutcome<User>>

const userResult = await fetchUser("user123")
if (userResult.isSuccess()) {
  console.log(userResult.value) // User object
}

// Convert TaskOutcome back to Promise (for interop)
const promise = Task.toPromise(asyncResult)
// Success → resolves with value
// Failure → rejects with error

Branded Types

import { Brand, ValidatedBrand } from "functype/branded"

// Create branded types for stronger type safety
type UserId = Brand<"UserId", string>
type Email = Brand<"Email", string>

// Simple branding - branded values ARE primitives!
const userId = Brand("UserId", "U123456")
console.log(userId) // "U123456" - it IS a string
console.log(typeof userId) // "string"
console.log(userId.toUpperCase()) // "U123456" - string methods work!

// Runtime-validated branding for safer input handling
const EmailValidator = ValidatedBrand("Email", (s: string) => /^[^@]+@[^@]+\.[^@]+$/.test(s))
const UserIdValidator = ValidatedBrand("UserId", (s: string) => /^U\d{6}$/.test(s))

// Safe creation with Option/Either return types
const email = EmailValidator.of("user@example.com") // Some(Brand<"Email", string>)
const invalidEmail = EmailValidator.of("invalid") // None

const userResult = UserIdValidator.from("U123456") // Right(Brand<"UserId", string>)
const userError = UserIdValidator.from("invalid") // Left("Invalid UserId: validation failed")

// Type safety in action
function getUserByEmail(email: Email): User {
  /* ... */
}

// These calls are type-safe
const userId = UserId("U123456")
const email = Email("user@example.com")
const user = getUserByEmail(email) // Works

// These would be type errors
getUserByEmail("invalid") // Type error: Argument of type 'string' is not assignable to parameter of type 'Email'
getUserByEmail(userId) // Type error: Argument of type 'UserId' is not assignable to parameter of type 'Email'

Conditional Programming

Functype provides Cond and Match for functional conditional logic without early returns:

Cond

import { Cond } from "functype"

// Replace if-else chains with Cond
const grade = Cond<number, string>()
  .case((score) => score >= 90, "A")
  .case((score) => score >= 80, "B")
  .case((score) => score >= 70, "C")
  .case((score) => score >= 60, "D")
  .default("F")

console.log(grade(85)) // "B"
console.log(grade(55)) // "F"

// With transformation
const discount = Cond<number, number>()
  .case(
    (qty) => qty >= 100,
    (qty) => qty * 0.2, // 20% off for 100+
  )
  .case(
    (qty) => qty >= 50,
    (qty) => qty * 0.1, // 10% off for 50+
  )
  .case(
    (qty) => qty >= 10,
    (qty) => qty * 0.05, // 5% off for 10+
  )
  .default(0)

console.log(discount(150)) // 30 (20% of 150)

Match

import { Match } from "functype"

// Pattern matching with Match
type Status = "pending" | "approved" | "rejected" | "cancelled"

const statusMessage = Match<Status, string>()
  .case("pending", "Your request is being processed")
  .case("approved", "Your request has been approved!")
  .case("rejected", "Sorry, your request was rejected")
  .case("cancelled", "Your request was cancelled")
  .exhaustive()

console.log(statusMessage("approved")) // "Your request has been approved!"

// Match with predicates
const numberType = Match<number, string>()
  .case(0, "zero")
  .case((n) => n > 0, "positive")
  .case((n) => n < 0, "negative")
  .exhaustive()

console.log(numberType(42)) // "positive"
console.log(numberType(-5)) // "negative"

Advanced Pattern Matching

Match supports exhaustive matching, nested patterns, and guards:

import { Match } from "functype"

// Exhaustive matching with compile-time checking
type Status = "idle" | "loading" | "success" | "error"
const result = Match<Status, string>("success")
  .case("idle", "Waiting...")
  .case("loading", "Loading...")
  .case("success", "Done!")
  .case("error", "Failed!")
  .exhaustive() // Compile error if any case is missing

// Nested pattern matching
type User = {
  name: string
  age: number
  role: "admin" | "user"
  preferences?: { theme: "light" | "dark" }
}

const message = Match<User, string>(user)
  .case({ role: "admin", age: (n) => n >= 18, preferences: { theme: "dark" } }, "Adult admin with dark mode")
  .case({ role: "user" }, (u) => `Regular user: ${u.name}`)
  .when((u) => u.age < 18, "Minor user - restricted access")
  .default("Unknown user type")

// Reusable pattern matchers
const classifier = Match.builder<Animal, string>()
  .when((a) => a.canFly, "Flying creature")
  .case({ legs: 0 }, "Legless")
  .case({ legs: 2 }, "Biped")
  .case({ legs: 4 }, "Quadruped")
  .default("Other")
  .build()

Fold

Functype includes a powerful fold operation for pattern matching and extracting values:

import { Option, Either, Try, List } from "functype"

// Option fold
const opt = Option(5)
const optResult = opt.fold(
  () => "None",
  (value) => `Some(${value})`,
) // "Some(5)"

// Either fold
const either = Right<string, number>(42)
const eitherResult = either.fold(
  (left) => `Left(${left})`,
  (right) => `Right(${right})`,
) // "Right(42)"

// Try fold
const tryValue = Try(() => 10)
const tryResult = tryValue.fold(
  (error) => `Error: ${error.message}`,
  (value) => `Success: ${value}`,
) // "Success: 10"

// List fold
const list = List([1, 2, 3])
const listResult = list.foldLeft(0)((acc, num) => acc + num) // 6

Foldable

Functype includes a Foldable type class that all data structures implement:

import { FoldableUtils, Option, List, Try } from "functype"

// All data structures implement the Foldable interface
const option = Option(5)
const list = List([1, 2, 3, 4, 5])
const tryVal = Try(() => 10)

// Use fold to pattern-match on data structures
option.fold(
  () => console.log("Empty option"),
  (value) => console.log(`Option value: ${value}`),
)

// Use foldLeft for left-associative operations
const sum = list.foldLeft(0)((acc, value) => acc + value) // 15

// Use foldRight for right-associative operations
const product = list.foldRight(1)((value, acc) => value * acc) // 120

// Use FoldableUtils to work with any Foldable
const isEmpty = FoldableUtils.isEmpty(option) // false
const size = FoldableUtils.size(list) // 5
const convertedToList = FoldableUtils.toList(option) // List([5])
const convertedToEither = FoldableUtils.toEither(tryVal, "Error") // Right(10)

Matchable

Functype includes a Matchable type class for enhanced pattern matching:

import { Option, Either, Try, List, MatchableUtils } from "functype"

// Pattern matching on Option
const opt = Option(42)
const optResult = opt.match({
  Some: (value) => `Found: ${value}`,
  None: () => "Not found",
}) // "Found: 42"

// Pattern matching on Either
const either = Either.fromNullable(null, "Missing value")
const eitherResult = either.match({
  Left: (error) => `Error: ${error}`,
  Right: (value) => `Value: ${value}`,
}) // "Error: Missing value"

// Pattern matching on Try
const tryVal = Try(() => JSON.parse('{"name":"John"}'))
const tryResult = tryVal.match({
  Success: (data) => `Name: ${data.name}`,
  Failure: (error) => `Parse error: ${error.message}`,
}) // "Name: John"

// Pattern matching on List
const list = List([1, 2, 3])
const listResult = list.match({
  NonEmpty: (values) => `Values: ${values.join(", ")}`,
  Empty: () => "No values",
}) // "Values: 1, 2, 3"

// Using MatchableUtils for advanced pattern matching
const isPositive = MatchableUtils.when(
  (n: number) => n > 0,
  (n) => `Positive: ${n}`,
)

const defaultCase = MatchableUtils.default((n: number) => `Default: ${n}`)

// Using pattern guards in custom matching logic
const num = 42
const result = isPositive(num) ?? defaultCase(num) // "Positive: 42"

Interface Hierarchy

All data structures in Functype implement a unified hierarchy of interfaces, providing consistent behavior across the library:

Type Classes

Functype leverages type classes to provide common operations:

• Functor: Supports map operation for transforming wrapped values
• Applicative: Extends Functor with ap for applying wrapped functions
• Monad: Extends Applicative with flatMap for chaining operations
• AsyncMonad: Extends Monad with flatMapAsync for async operations
• ContainerOps: Universal operations for all containers (single-value and collections)
• CollectionOps: Operations specific to collections like List and Set

Unified Interfaces

All data structures implement the Functype hierarchy:

// Base interface for all data structures
interface FunctypeBase<A, Tag>
  extends AsyncMonad<A>, Traversable<A>, Serializable<A>, Foldable<A>, Typeable<Tag>, ContainerOps<A> {
  readonly _tag: Tag
}

// For single-value containers (Option, Either, Try)
interface Functype<A, Tag> extends FunctypeBase<A, Tag>, Extractable<A>, Pipe<A>, Matchable<A, Tag> {
  toValue(): { _tag: Tag; value: A }
}

// For collections (List, Set, Map)
interface FunctypeCollection<A, Tag>
  extends FunctypeBase<A, Tag>, Iterable<A>, Pipe<A[]>, Collection<A>, CollectionOps<A, FunctypeCollection<A, Tag>> {
  toValue(): { _tag: Tag; value: A[] }
  // Collections work with Iterable instead of Monad
  flatMap<B>(f: (value: A) => Iterable<B>): FunctypeCollection<B, Tag>
}

Container Operations

All containers (Option, Either, Try, List, Set) support these universal operations:

import { Option, List } from "functype"

const opt = Option(42)
const list = List([1, 2, 3, 4, 5])

// Universal operations work on both single-value and collections
opt.count((x) => x > 40) // 1
list.count((x) => x > 3) // 2

opt.find((x) => x > 40) // Some(42)
list.find((x) => x > 3) // Some(4)

opt.exists((x) => x === 42) // true
list.exists((x) => x === 3) // true

opt.forEach(console.log) // Logs: 42
list.forEach(console.log) // Logs: 1, 2, 3, 4, 5

Feature Matrix

For a comprehensive overview of which interfaces are supported by each data structure, see the Functype Feature Matrix.

Type Safety

Functype leverages TypeScript's advanced type system to provide compile-time safety for functional patterns, ensuring that your code is both robust and maintainable.

// Type inference works seamlessly
const option = Option(42)
// Inferred as number
const mappedValue = option.map((x) => x.toString())
// Inferred as string

Error Formatting

Functype provides utilities for improved error visualization and logging:

import { formatError, createErrorSerializer } from "functype/error"

// Create a nested task error
const innerTask = Task({ name: "DbQuery" }).Sync(() => {
  throw new Error("Database connection failed")
})

const outerTask = Task({ name: "UserFetch" }).Sync(() => {
  return innerTask.value
})

// Format the error for console display
console.error(
  formatError(outerTask.value as Error, {
    includeTasks: true,
    includeStackTrace: true,
    colors: true,
  }),
)

// Create a serializer for structured logging libraries like Pino
const errorSerializer = createErrorSerializer()

// Use with Pino
const logger = pino({
  serializers: { err: errorSerializer },
})

// Log the error with full context
logger.error(
  {
    err: outerTask.value,
    requestId: "req-123",
  },
  "Failed to fetch user data",
)

For more details, see the Error Formatting Guide.

Roadmap / TODO

High Priority

• Complete LazyList Implementation
  • ✓ Add Foldable interface (fold, foldLeft, foldRight)
  • ✓ Add Pipe interface for composition
  • ✓ Add Serializable for persistence
  • ✓ Add Typeable support
• Implement NonEmptyList
  • List guaranteed to have at least one element
  • Prevents empty list errors at compile time
  • Full standard interface implementation
  • Methods like head return A instead of Option<A>

Medium Priority

• Implement ValidatedNel<E, A> for validation with error accumulation
  • Unlike Either, collects multiple errors
  • Uses NonEmptyList for error collection
  • Applicative instance combines errors
• Enhance Pattern Matching
  • ✓ Add exhaustiveness checking at compile time
  • ✓ Support nested pattern matching
  • ✓ Add guard clauses (when conditions)
  • ✓ Support destructuring patterns
  • ✓ Consolidated into unified Match implementation
• Implement IO<R,E,A> effect type
  • ✓ Lazy execution with typed errors
  • ✓ Dependency injection (Tag, Context, Layer)
  • ✓ Composable operations (map, flatMap, zip, race)
  • ✓ Resource management (bracket/acquireRelease)
  • ✓ Generator do-notation (IO.gen) and builder (IO.Do)

Low Priority

• Complete Tuple Implementation
  • ✓ Add Foldable for tuple operations
  • ✓ Add Pipe interface for composition
  • ✓ Add Serializable for persistence
  • ✓ Add Companion pattern with utility methods
  • ✓ Added specialized pair() and triple() constructors
• Implement Lens<S, A> for immutable updates
  • Composable property access
  • Type-safe nested updates
  • Works with all functype data structures
• Add Reader/State monads for dependency injection and state management

Completed Functionality

• Add lazy evaluation structures (LazyList implemented, needs interface completion)
• Add a proper Foldable type class interface
• Implement Matchable type class for pattern matching
• Implement Applicative and other functional type classes (for most types)

Performance Optimizations

• Add memoization utilities (via Lazy — memoized deferred computation)
• Improve recursive operations for large collections
• Implement immutable data structures with structural sharing
• Add performance benchmarks (bench/ — Do-notation, monad detection, optimization targets)
• Optimize TreeShaking with sideEffects flag in package.json
• Support selective module imports for smaller bundles
• Add bundle size monitoring to CI/CD

API Consistency

• Ensure all modules follow the Scala-inspired pattern:
  • Constructor functions that return objects with methods
  • Object methods for common operations
  • Companion functions for additional utilities
• Align Task API with other monadic structures
• Standardize import patterns (@ imports vs relative paths)
• Implement consistent error handling strategy for async operations

Testing and Documentation

• Add observable test coverage metrics
• Implement property-based testing
• Expand error handling tests (test/error/ — typed errors, pino integration, stack traces)
• Add interoperability tests with other libraries

TypeScript Improvements

• Enable stricter TypeScript settings (noImplicitAny: true)
• Add noUncheckedIndexedAccess for safer array indexing
• Improve support for higher-kinded types:
  • Current type parameters work well for first-order types
  • Expand to support type constructors as parameters (F => F)
• Add branded/nominal types for stronger type safety
• Implement more type-level utilities — partial (Widen<A, B> conditional-type helper in src/typeclass/variance.ts; more in src/types/, src/typeable/)
• Leverage newer TypeScript features (const type parameters, tuple manipulation)

Completed Major Features (0.60.0 audit)

Work that shipped but wasn't captured in the planned roadmap above:

• Scala-aligned covariance for the container family (0.57.2 → 0.60.0)
  • ✓ <out T> / <out K, out V> across Option, Either, Try, List, Set, Map, Lazy, LazyList, Tuple, Identity, Task/TaskOutcome
  • ✓ FunctypeSum hierarchy separating sum types (Either, Try) from iterables
  • ✓ Widen<A, B> conditional-type helper enforcing Scala's [B >: A] lower-bound semantics in reduce / reduceRight
  • ✓ Method-level widening patterns (or, orElse, recover, add, concat, etc.)
  • ✓ Contributor guide at docs/variance-guide.md
  • Deferred: ZIO-style <in R, out E, out A> for IO
• Do-notation with generator comprehensions (src/do/)
• HTTP module — typed fetch wrapper with validator support (src/fetch/ — Http, HttpClient, HttpError, HttpRequest)
• Throwable — enhanced error wrapper preserving stack traces and context (src/core/throwable/)
• Task / TaskOutcome — sync/async with cancellation + progress tracking (src/core/task/)
• IO effect system companions — Context, Tag, Layer, Exit, TestClock (src/io/)
• Ref — mutable reference for controlled mutability (src/ref/)
• Identity — trivial covariant container (src/identity/)
• Conditional module (src/conditional/)
• Valuable module (src/valuable/)
• MCP server for functype documentation (mcp-server/)
• CLI — npx functype (src/cli/)
• Claude Code skills — functype (user guide) and functype-developer (contributor guide)
• Serialization infrastructure — JSON / YAML / binary (src/serialization/)

Claude Code Skills

Functype provides two specialized Claude Code skills to enhance your development experience:

Functype User Guide

Help for developers using functype in their projects:

• Pattern conversion (imperative → functional)
• API lookup and examples
• Common use cases and debugging tips

Install: Download from dist/skills/functype-user.zip

Functype Library Developer

Guide for contributors developing functype itself:

• Architecture patterns and development workflow
• Creating new data structures
• Testing strategies and interface implementation

Install: Download from dist/skills/functype-developer.zip

See .claude/skills/README.md for complete installation and usage instructions.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT License

Copyright (c) 2025 Jordan Burke

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.