Mastering Functional Programming with TypeScript
A practical guide to functional programming in TypeScript using the fp-ts library, covering core concepts like referential transparency, currying, functors, and monads.
Core Framework
Functional programming treats functions as mappings between sets. Pure functions have referential transparency: when a function's result is known for given parameters, you can replace the call with the result. Side effects break this property.
Unlike object-oriented programming where methods attach to data, functional programming keeps code and data separate. Both are first-class citizens.
Key Concepts
Currying and Composition
Currying transforms multi-argument functions into chains of single-argument functions. Named after mathematician Haskell Curry (whose first name inspired the Haskell language).
- pipe: Takes data as the first parameter, then chains functions that transform it
- flow: Like pipe without the initial data, returning a composed function
Both are higher-order functions.
Algebraic Data Types
- Functor: Any container holding values that exposes a
mapfunction to transform those values while preserving the container type - Monoid: Knows how to
concatand has an empty element - Monad: Unlocks Do notation, allowing side computations alongside the main computation
Spreading objects into new ones replaces inheritance. You can compose from multiple sources, not just a single ancestor chain.
Side Effects and IO
A side effect is any change to the execution environment peripheral to the main computation. Marking functions as IO doesn't remove side effects; it lets you compose effectful computations as if they were pure.
fp-ts Approach
Native JavaScript favors method cascading: arr.map(func).reduce(func). fp-ts prefers piped invocations. Pipe detaches logic from the subject.
Practical Guidance
Avoid explicit for/while loops or abrupt terminations like throwing exceptions. Reaching for these signals a less-optimal path; step back and reconsider the solution.