From mboxrd@z Thu Jan  1 00:00:00 1970
Received: (from weis@localhost) by pauillac.inria.fr (8.7.6/8.7.3) id TAA26412 for caml-redistribution; Wed, 19 May 1999 19:45:18 +0200 (MET DST)
Received: from nez-perce.inria.fr (nez-perce.inria.fr [192.93.2.78]) by pauillac.inria.fr (8.7.6/8.7.3) with ESMTP id BAA11071 for <caml-list@pauillac.inria.fr>; Tue, 18 May 1999 01:55:21 +0200 (MET DST)
Received: from graphics.lcs.mit.edu (graphics.lcs.mit.edu [18.24.2.30])
	by nez-perce.inria.fr (8.8.7/8.8.7) with ESMTP id BAA16929
	for <caml-list@inria.fr>; Tue, 18 May 1999 01:55:20 +0200 (MET DST)
Received: from turpentine.lcs.mit.edu (turpentine.lcs.mit.edu [18.24.2.91])
	by graphics.lcs.mit.edu (8.9.0/8.9.0) with ESMTP id TAA14156
	for <caml-list@inria.fr>; Mon, 17 May 1999 19:53:40 -0400 (EDT)
Received: from localhost (b-db@localhost)
	by turpentine.lcs.mit.edu (8.9.0/8.9.0) with SMTP id TAA40711
	for <caml-list@inria.fr>; Mon, 17 May 1999 19:53:39 -0400 (EDT)
Date: Mon, 17 May 1999 19:53:39 -0400 (EDT)
From: Benoit deBoursetty <b-db@graphics.lcs.mit.edu>
Reply-To: Benoit deBoursetty <b-db@graphics.lcs.mit.edu>
To: Caml mailing list <caml-list@inria.fr>
Subject: modification to the O'CaML sets.ml
In-Reply-To: <199905141108.NAA02693@pauillac.inria.fr>
Message-ID: <Pine.SGI.3.93.990517183050.59952A-100000@turpentine.lcs.mit.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: weis

Those who need the 1-type-variable polymorphic set data structure I've 
talked of recently will find this file useful :

http://www.polytechnique.fr/eleves/1996/boursetty/polyset.ml

!! Warning !! The URL will stop being effective at the beginning of July.
Would anyone download this very small file to a more persistent website ?

It is a small modification to the original set.ml

In fact, only type declarations are modified, the code is just the same. 
Doesn't this mean that the compiler polymorphism handling could be
enhanced ? 

In the case of sets I feel like declaring something like

(* This is not valid o'caml *)
module type 'a Comparison =
  sig
    val compare : 'a -> 'a -> int
  end

and a functor from modules with type 'a Comparison to modules with type

(* Neither is this *)
module type 'a Set =
  sig
    type 'a t
    val empty: 'a t
    val is_empty: 'a t
    val mem: 'a -> 'a t -> bool
    val add: 'a -> 'a t -> 'a t
  .
  .
  .
  end

That is only my personal, programmer's point of view. I don't know of the
typing problems behind it.

Benoit de Boursetty