Uploaded by
Published on

This library provides a newtype over Int which can represent the ring of integers modulo n for any positive integer n, sometimes written ℤ/nℤ.

If you haven't heard of this before, we will use the integers modulo 12 as an example. There are 12 distinct elements of this set: 0, 1, 2, ... 9, 10, 11. (In general, there are n elements; 0 up to n-1). Addition works as normal, except that if the result is larger than 12, we loop around to 0 again so that we stay in the same set. For example, with integers modulo 12, 2 + 3 = 5 (as normal), but 10 + 5 = 3. Multiplication works in a similar way, looping back around if the result would be too large so for example 3 * 6 = 6.

What's the point of this library though? Well, we do often want to deal with integers modulo n in 'real code'. For instance, we might use integers modulo n to represent player IDs in a turn-based game of n players: then, the function for selecting the next player to take a turn is simply (_ + one); this will then automatically loop back to 0 after all the players have taken a turn.

Another application of this library is for testing libraries which abstract over the numeric type classes defined in the Prelude. Int and Number are often not suitable for testing with, because they do not always abide by the relevant type class laws. By comparison, the types provided by this library are always fully law-abiding (well, as long as you don't ask for integers modulo some number larger than 2^31).

In fact, you even get a (fully law-abiding) Field when n is a prime number! We can divide elements of ℤ/nℤ when n is prime by finding the multiplicative inverse of the divisor: that is, the number that you have to multiply it by to get 1. Then, x divided by y is the same as x times the inverse of y. The prime number restriction is required to ensure that every number has a unique inverse. So for example, in ℤ/7ℤ, the multiplicative inverse of 5 is 3, because 5 * 3 = 1. Therefore, 4 / 5 = 4 * 3 = 5. This still behaves how we would hope, since multiplying by 5 again gets us back to where we started: 5 * 5 = 4.

Documentation is on Pursuit.