Type class confusion

[zevv]

I'm not even sure what a proper title for this would be, sorry.

I am not sure if this is a bug or if I understand the manual wrong. The following fails to compile:

type IntOrFloat = int | float

proc foo(a: IntOrFloat, b: IntOrFloat) =
  discard

foo(3.int, 3.int)      # ok
foo(3.float, 3.float)  # also ok

foo(3.float, 3.int)    # type mismatch!
foo(3.int, 3.float)    # type mismatch!

The error being:

Error: type mismatch: got <float, int>
but expected one of: 
proc foo(a: IntOrFloat; b: IntOrFloat)
  first type mismatch at position: 2
  required type for b: IntOrFloat
  but expression 'int(3)' is of type: int

However, if I change the proc signature to his, things are just fine:

proc foo[A, B: IntOrFloat](a: A, b: B) =
  discard

Also works:

proc foo(a: int | float, b: int | float) =
  discard

[jrfondren]

@zevv what conclusion did you come to about this?

[zevv]

That I was not smart enough to properly read the manual.

https://nim-lang.github.io/Nim/manual.html#generics-type-classes

"Whilst the syntax of type classes appears to resemble that of ADTs/algebraic data types in ML-like languages, it should be understood that type classes are static constraints to be enforced at type instantiations. Type classes are not really types in themselves, but are instead a system of providing generic "checks" that ultimately resolve to some singular type. Type classes do not allow for runtime type dynamism, unlike object variants or methods"