functional programming - bencope.land (Page 2)

bencope.land

functional programming

A collection of 40 posts

functional programming

Lambda Calculus and Arguments

Functions in the lambda calculus are always unary - they only take one argument. Functional programming is based on the lambda calculus, and it is useful to learn to appreciate unary functions. One useful way to train your brain to think functionally is, when you find yourself writing an n-ary

Type-safe JavaScript Development with F# and Fable

F# is a suitable alternative to TypeScript for writing a JavaScript application with type-safety. But you also get the additional benefits of F#'s powerful type inference. TypeScript interface Person { firstName: string lastName: string } let people: Person[] = [ { firstName: 'John', lastName: 'Doe' }, { firstName: 'Jane'

functional programming

FP Lint

ESLint can help you enforce functional-first standards. Below is a functional eslint setup I like, followed by some explaination. { "root": true, "parser": "@typescript-eslint/parser", "plugins": [ "@typescript-eslint", "fp", "prefer-let" ], "extends": [ "airbnb", "plugin:

game development

Preferring Immutability in Godot

Game engine development tends to rely on mutability, such as keeping up with an object's position on-screen. Typically, the game object is represented by a single class with a "position" property, and as the object moves around that property mutates to reflect its new value. This

game development

Exporting Godot Variables in F#

In Godot, it's convenient to export variables so that they can be updated via the Inspector tab. Doing so in C# looks like this: using Godot; public class Player : Area2D { [Export] public float Speed = 2f; public override void _Ready() {} public override void _Process(float delta) {} } Here the Speed

lambda calculus

Argument Selection Function

The argument selection function λfirst.λsecond.first accepts a function as an argument and returns a function which, when given an argument, returns the first argument. // λfirst.λsecond.first const selectFirst = x => y => x const double = a => a * 2 const triple = a => a * 3 const applyDouble

functional programming

Semantic Focus

In his book Stylish F#, Kit Eason describes a few unstylish pitfalls that developers may fall into. One of these pitfalls he calls lack of semantic focus. I'll describe the kind of code that isn't readable with minimum context as having poor semantic focus. In other

functional programming

Functional Reactive Programming

There's something about FRP specifically that makes programming an absolute pleasure. Designing an elegant reactive application is a great feeling and opens you're mind to just how rewarding it can be to focus on decoupling. Thinking in reactive (message driven) streams has clear benefits between multiple

game development

Godot and F#

Over the weekend I got a game started with Godot Mono and F#. It works out pretty well, although it relies on using C# to extend types defined in F#. Out of the gate I've taken up a convention of defining a namespace just for defining a type

lambda calculus

Application Function

Function application function is represented by the term λf.λx.f x. It accepts a function as an argument and applies it to the argument of the function it returns. // λf.λx.f x let apply f x = f x let addOne c = c + 1 let applyAddOne = addOne |> apply

lambda calculus

Self Application Function

The function λx.x x is the self application function. It applies its argument to its argument. // λx.x x const sa = x => x(x) Like the identity function, it doesn't seem helpful at first glance but it has some interesting applications. Consider the twice function λf.

lambda calculus

Identity Function

Identity function is the simplest of the lambda calculus functions, represented by λa.a. It simply returns the argument that it was given. // F# fun a -> a // JS a => a # Python lambda a: a While it is a simple function, it still has significance in composing functions

functional programming

Pipes Again

Python's PyMonad library provides some neat utilities for functional programming. Here's a Python pipe function from a previous post. pipe = lambda *fns: lambda x: reduce(lambda y, f: f(y), list(fns), x) g = lambda n: n + 1 f = lambda n: n * 2 h = pipe(g,

functional programming

Pipes

Inspired by the following Tweet from Eric Elliot in which he uses a reducer to make a pipe function, I thought I'd do something similar in Python and F#. const pipe = (...fns) => x => fns.reduce((y, f) => f(y), x); const g = n => n

Deno, Node and TypeScript

JavaScript is considered decent language for functional programming, particularly because its weak, dynamic typing combined with arrow functions make for succinct code. After using F# more though I've really come to appreciate type-driven development. Strong types and functional programming go so well together, I've essentially completely