Question

I want to write a function which operates on double and any other type of number supporting multiplication and addition, yielding double as a result.
The following, of course, doesn’t compile, since type of (*) is t -> t -> t, so mixing different types is not allowed:

f :: (Num a) => Double -> a -> a -> Double
f x a b = a*x + b

What I want is the ability to write something like this:

f :: ...
f x a b = ... -- equivalent to a*x + b

f 1.0 (2 :: Int)    (3 :: Int)    -- returns 5.0
f 1.0 (2 :: Word32) (3 :: Word32) -- returns 5.0
f 1.0 (2 :: Float)  (3 :: Float)  -- returns 5.0

What should I do to make it work? Or maybe I’m fundamentally wrong and shouldn’t be doing this? It is very strange but I didn’t find anything on the internet about this.

Answer 1

In RWH.chapter6 there is a nice paragraph about converting numbers between some numeric types (Table 6.4).

f :: (Real a) => Double -> a -> a -> Double
f x a b = x * (cast a) + (cast b)
  where cast = fromRational . toRational

Seems workable.

> f 1.0 (2 :: Int)    (3 :: Int)
5.0
it :: Double
> f 1.0 (2 :: Word32) (3 :: Word32)
5.0
it :: Double
> f 1.0 (2 :: Float)  (3 :: Float)
5.0
it :: Double