Re: [Caml-list] Integer exponentiation and **

[Jacques-Henri Jourdan <jacques-henri.jourdan@normalesup.org>]

Hi,

Concerning integer exponentiation: integers in OCaml are bounded, and 
not meant to compute over large integers. This means that exponentiation 
will in practice overflow very quickly, and, anyway, there is no 
hardware support for exponentiation (meaning that it would be slower 
than expected). There are two alternatives : if the base is 2, you can 
use bitwise shifts, and, in any case, you can use a large integers 
library such as ZArith (1).

Concerning the syntax for floating point exponentiation: I agree that it 
is not coherent. I guess that's a matter of taste (as always with 
syntax): some prefer typing 2 characters instead of 3. I honestly do not 
know if there is any other reason.You can still define the other syntax:

     let ( **. ) = ( ** );;

-- 
JH

(1) https://github.com/ocaml/Zarith

On 01/24/2017 09:23 AM, Yann Salmon wrote:
> Hello,
>
> I am wondering why OCaml does not have an integer exponentiation
> operator and why the floating point exponentiation operator is ( ** )
> and not ( **. ).
>
> Any ideas ?
>