From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: weis
Received: (from weis@localhost) by pauillac.inria.fr (8.7.6/8.7.3) id KAA21284 for caml-redist@pauillac.inria.fr; Tue, 16 May 2000 10:21:41 +0200 (MET DST)
Resent-Message-Id: <200005160821.KAA21284@pauillac.inria.fr>
Received: from concorde.inria.fr (concorde.inria.fr [192.93.2.39]) by pauillac.inria.fr (8.7.6/8.7.3) with ESMTP id AAA05659 for <caml-list@pauillac.inria.fr>; Tue, 16 May 2000 00:20:59 +0200 (MET DST)
Received: from sarg.Ryerson.CA (sarg.Ryerson.CA [141.117.18.117])
	by concorde.inria.fr (8.10.0/8.10.0) with ESMTP id e4FMKwT23416
	for <caml-list@inria.fr>; Tue, 16 May 2000 00:20:58 +0200 (MET DST)
Received: from sarg.Ryerson.CA (dmason@localhost [127.0.0.1])
	by sarg.Ryerson.CA (8.8.7/8.8.7) with ESMTP id SAA07498
	for <caml-list@inria.fr>; Mon, 15 May 2000 18:20:44 -0400
Message-Id: <200005152220.SAA07498@sarg.Ryerson.CA>
To: caml-list@inria.fr
Subject: Re: reference initialization 
In-reply-to: Your message of "Mon, 15 May 2000 13:47:15 EDT."
             <Pine.GSO.4.21.0005151309280.9542-100000@gatekeeper.ececs.uc.edu> 
Mime-Version: 1.0 (generated by tm-edit 7.108)
Content-Type: text/plain; charset=US-ASCII
Date: Mon, 15 May 2000 18:20:44 -0400
From: Dave Mason <dmason@sarg.Ryerson.CA>
Resent-From: weis@pauillac.inria.fr
Resent-Date: Tue, 16 May 2000 10:21:41 +0200
Resent-To: caml-redist@pauillac.inria.fr

>>>>> On Mon, 15 May 2000 13:47:15 -0400 (EDT), Hongwei  Xi <hwxi@ECECS.UC.EDU> said:


> What I had in mind is actually something pretty simple (since it
> won't be very useful if it is not simple :-))

> Is it possible to have something like the following in the library:

I propose something more like:

    Array.init_with_array: int -> (int -> (int -> 'a) -> 'a) -> 'a Array

let Array.init_with_array n f =
  let a = Array.create n in
    for i = 0 to n - 1 do
      a.(i) <- f i (fun j -> if j<i then
                                a.(i)
                             else raise ``initialization error'')
    done

The second parameter could instead be a version of the array but with
the bound limited to 0..i, but this would allow the function f to save
that version somewhere which might not be good.

This would solve an additional subset of the cases... but is NOT a
complete solution (personally, I'm fine with the initial value).

../Dave