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 TAA24827 for caml-redistribution; Wed, 4 Jun 1997 19:26:37 +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 KAA15519 for <caml-list@pauillac.inria.fr>; Wed, 4 Jun 1997 10:34:01 +0200 (MET DST)
Received: from arthur.u-strasbg.fr (boos@arthur.u-strasbg.fr [130.79.6.103]) by nez-perce.inria.fr (8.8.5/8.7.3) with ESMTP id KAA02625 for <caml-list@inria.fr>; Wed, 4 Jun 1997 10:33:58 +0200 (MET DST)
Received: (from boos@localhost) by arthur.u-strasbg.fr (8.7.5/8.7.3) id MAA15000; Wed, 4 Jun 1997 12:35:41 +0200
Date: Wed, 4 Jun 1997 12:35:41 +0200
Message-Id: <199706041035.MAA15000@arthur.u-strasbg.fr>
From: Christian Boos <boos@arthur.u-strasbg.fr>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: caml-list@inria.fr
Subject: partial instantiation of a new obj. : bug or feature ?
X-Mailer: VM 6.22 under 19.15p4 XEmacs Lucid
Sender: weis


Hi, 

It seems that partial instantiation of objects' new statement didn't work 
as I expected.
I thought that the creation of an object takes place when all the arguments
are available.

The following example illustrates this:

  # class test (a : int) (b : int) =
       val a = a
       val b = b
       method get = (a, b)
    end;;
  class test (int) (int) = val a : int val b : int method get : int * int end
  # let a1 = new test 1;;
  val a1 : int -> test = <fun>
  # let b1 = a1 1;;
  val b1 : test = <obj>
  # let b2 = a1 2;;
  val b2 : test = <obj>
  # b1 = b2;;
  - : bool = true
  # b1#get;;
  - : int * int = 1, 2
  # b2#get;;
  - : int * int = 1, 2
  # 

This forces one to write things like :

  let objs = Array.map (fun i -> new test 1 i) vals

instead of the more intuitive :

  let objs = Array.map (new test 1) vals


Maybe someone wants to comment ?

-- Christian