ddd-ts-fp

DDD Functional Patterns

Immutable Aggregate

Model aggregates as immutable types with pure curried functions, not classes with methods:

// Immutable type + curried functions
type Receipt = {
  readonly id: ReceiptId;
  readonly tenantId: TenantId;
  readonly addresses: readonly Address[];
  readonly period: Period;
  readonly amount: number;
  readonly createdAt: Date;
};

const addAddress =
  (address: Address) =>
  (receipt: Receipt): Receipt => ({
    ...receipt,
    addresses: [...receipt.addresses, address],
  });

Each operation is a curried function that returns a new aggregate (or a Result). Compose with pipe and chain.

When implementing aggregate operations, pipe composition, or nested immutable updates, read references/ddd-functional-examples.md for complete patterns with pure and fallible transformations.

Smart Constructor

make prefix: curried factory function that captures context and returns a specialized function, usable directly in a pipe:

const makeAddress =
  (addressId: string, createdAt: Date) =>
  (command: AddAddressCommand): Result<Address, DomainError> =>
    ok({ id: addressId, street: command.street, city: command.city, ... });

Rule	Convention
Prefix	make
Signature	make*(context) => (input) => output
Return	Often Result<T, E>, sometimes T directly
Position	End of pipeline (after validations) or start of workflow

When creating smart constructors for domain objects, read references/ddd-functional-examples.md for complete examples and pipeline integration.

Validation Pipeline

Validate incoming data via a chain of composable validators. Stops at first error:

type Validator<T> = (input: T) => Result<T, DomainError>;

// Composition
const validateCreateReceipt = (cmd) =>
  pipe(cmd, validatePeriod, chain(validateAmount), chain(validateLease));

Enrichment Pipeline

Transform external data into domain objects via progressive enrichment. Each step enriches or transforms the command.

Aspect	Validation	Enrichment
Purpose	Verify constraints	Transform / complete data
Source	Internal data (domain)	External data (API, message broker)
Position	Before business logic	At system boundary

When building validation or enrichment pipelines, read references/ddd-functional-examples.md for complete pipeline implementations with enrichers.

Handler Pattern

The handler orchestrates: retrieval, validation, transformation, persistence.

Flow: validate -> load aggregate -> domain logic (pure) -> persist.

When writing domain handlers, read references/ddd-functional-examples.md for the complete handler pattern with error handling.

Quick Reference

Rule	Convention
Aggregate	Immutable readonly type, no class
Operations	Pure curried functions
Smart Constructor	make*(context) => (input) => output
Updates	Spread operator, never mutate
Composition	pipe(aggregate, op1, op2, op3)
Fallible operations	pipe(aggregate, op1, chain(op2))
Validation	Composable Validator<T> pipeline
Enrichment	Pipeline at system boundary
Handler	Orchestrator: validate -> load -> domain -> persist