indique/elm-n-nat/4.0.1

elm-n-nat

Current attempts of represening natural numbers (>= 0) only have quite lose promises.

This package contains a very strict Nat type. The actual value is present in the type.

If you only want to ensure that a Nat is within a minimum (& maximum), elm-nat-in-range is the right choice!

Setup

elm install indique/elm-n-type
elm install indique/elm-n-nat-type
elm install indique/elm-n-nat

import NNat exposing (NNat)
import N.Type exposing (..) --N, Is, Difference, To
import N.Nat.Type exposing (..) --Nat0 to Nat192

Lets take a look at its types

n0 :
    NNat --a natural number
        (N Nat0 --zero is the exact value at compile time
         Is (Difference a To a)
         --we can also describe zero as the difference a - a
        )

sub1 :
    NNat
        (N (Nat1Plus nMinus1)
            --the exact value is nMinus1 + 1
            Is (Difference a To (Nat1Plus nPlusAMinus1))
            --we can also describe it as the difference (a + nMinus1 + 1) - a
        )
        
    -> NNat (N nMinus1 Is (Difference a To nPlusAMinus1))

→ sub1 n0 is a compile-time-error!

→ We can only call sub12 for a NNat that is at least 12

Most of the time, you rather want a nat-in-range to do calculations.

You can also express things like

• one NNat should be at least as big as another one.

  safeInterval :
      NNat (N first Is (Difference lastMinusFirst To last))
      -> NNat (N last Is lastDifference)
      -> Interval

Take a look at elm-nat-in-range & elm-array-in-range to see how NNat is used!