What is the workaround for OCaml: exception Invalid_argument("Random.int")?

Question

I have this bit of code:

let rec random_list = function
  | 0 -> []
  | n -> ( Random.int max_int ) :: ( random_list (n-1) )

It compiles okay, but when I execute it, this error shows up:

exception Invalid_argument("Random.int")

What is the workaround for this issue ?

Théo Winterhalter · Answer 1 · 2015-06-17T07:26:08.013

11

The documentation says:

Random.int bound returns a random integer between 0 (inclusive) and bound (exclusive). bound must be greater than 0 and less than 2^30.

So the closest to what you want is:

let my_max_int = (1 lsl 30) - 1 in
Random.int my_max_int

As gsg suggested, using Random.bits () is cleaner to get almost the same result (it can also return 2^30 - 1).

If you really want to get any positive native integer, maybe you could use Random.nativeint, but this means you would have to use the module Nativeint and Nativeint.t instead of int.

edited Jun 17 '15 at 07:26

answered Jun 16 '15 at 12:48

Théo Winterhalter

4,908
2
18
34

Thank you for your help. – Stefan Ciprian Hotoleanu Jun 16 '15 at 12:52
You are welcome. I improved my answer with a way to write `2^30 - 1` in an effective way. – Théo Winterhalter Jun 16 '15 at 12:55
1

`Random.bits ()` is a cleaner way to do this. – gsg Jun 16 '15 at 13:07
Oh yes, I will add it to my answer then. – Théo Winterhalter Jun 16 '15 at 13:09
2

`Random.bits ()` is not the same as the above, because `Random.bits ()` generates a random 30-bit positive integer, which can produce 2^30-1, whereas `Random.int ((1 lsl 30) - 1)` cannot produce 2^30-1. – newacct Jun 16 '15 at 22:44
My experience to share: I tried `Random.bits()` and `let my_max_int = (1 lsl 30) - 1 ` and got `stackoverflow`. I changed it to `let my_max_int = (1 lsl 25) - 1 ` and the error stops. – armand Jun 11 '21 at 19:18

What is the workaround for OCaml: exception Invalid_argument("Random.int")?

1 Answers1