F# vs C#

Every year I dabble with F# for several weeks. I did this again during the 2016 holiday break. Before I forget what I learned this break I thought it would be good to recap the major benefits of F# over C#.

Better Default Equality

Let's start by defining a Point model in C#. One might make a struct like this:

public struct Point
{
    public float X { get; set; }
    public float Y { get; set; }
}

In F# you can get the above model with the equality property in much simpler way like this:

type Point = { X:float; Y:float}

In F#, types have built in structural equality and comparison.

Better Compile-Time Checks

F# has a better way of enforcing you cover all cases during compile time. If you make a shape type for squares, rectangles, and circles you could model it like this:

type Shape =
    | Square of float
    | Rectangle of float * float
    | Circle of float

If you were then to go and implement an area function you would implement it like this:

let Area (shape: Shape) =
    match shape with 
    | Square x -> x * x
    | Rectangle(h,w) -> h*w
    | Circle r -> System.Math.PI * r *r

The compiler forces you to match on Square, Rectangle, and Circle.

Better Handling of NULLs

In C# given a method like this:

public List<string> GetTopPlayers(List<Players> players, int n) 
{ ... }

You cannot easily determine what you will get back if you pass in an empty list of players. You might get back a null, an empty list, or an InvalidArgumentException could be thrown.

F# can return what's called "None" and the caller has to handle the "None" case.