From mboxrd@z Thu Jan  1 00:00:00 1970
Received: (from majordomo@localhost) by pauillac.inria.fr (8.7.6/8.7.3) id UAA01981; Thu, 5 Apr 2001 20:15:42 +0200 (MET DST)
X-Authentication-Warning: pauillac.inria.fr: majordomo set sender to owner-caml-list@pauillac.inria.fr using -f
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 UAA01732 for <caml-list@pauillac.inria.fr>; Thu, 5 Apr 2001 20:15:41 +0200 (MET DST)
Received: from shell5.ba.best.com (shell5.ba.best.com [206.184.139.136])
	by nez-perce.inria.fr (8.11.1/8.10.0) with ESMTP id f35IFe502754
	for <caml-list@inria.fr>; Thu, 5 Apr 2001 20:15:40 +0200 (MET DST)
Received: from localhost (bpr@localhost)
	by shell5.ba.best.com (8.9.3/8.9.2/best.sh) with ESMTP id LAA05341;
	Thu, 5 Apr 2001 11:15:37 -0700 (PDT)
Date: Thu, 5 Apr 2001 11:15:37 -0700 (PDT)
From: Brian Rogoff <bpr@best.com>
To: Nobuyuki Tomizawa <tomizawa@ccm.cl.nec.co.jp>
cc: caml-list@inria.fr
Subject: Re: [Caml-list] Question: 'instanceof'-like like primitive in OCaml
In-Reply-To: <FNEDLLDONJLKENCGLOOIKENGCJAA.tomizawa@ccm.cl.nec.co.jp>
Message-ID: <Pine.BSF.4.21.0104051058580.18792-100000@shell5.ba.best.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-caml-list@pauillac.inria.fr
Precedence: bulk

On Wed, 4 Apr 2001, Nobuyuki Tomizawa wrote:
> Dear all,
> 
> I'm a novice OCaml programmer and have a question about heterogeneous
> list and "downward cast".

There are no downcasts in the OCaml object system. Someone posted a patch
which would allow it but if you want to use standard OCaml you have to
write code without it. It is bad OO style anyways, right? 

> and what I want to do are:
> 
>      * make the list which typed as "base list".
>      * call `derived2#specific' method if the element in the list is
>      an instance of 'derived2'.
> 
> But, OCaml seems not to have Java's `instanceof'-like primitive and/or
> downward-cast primitive.
> 
> My solution is to use variant type for the list and identify the class
> using pattern matching:
> 
>     type tag = Derived2 of d2 | DontCare of b;;
> 
>     let l = [ Derived2(new d2); DontCare(new d1 :> b)] in ...;;
> 
> But I feel this solution is awkward because we have to define variant
> type for each classes I want to treat them as specific.
> 
> Could you please tell me more 'smart' answer or another way in OCaml
> style?

Actually, what you have isn't too bad, and you have to use variants to
get the capability that you want. I do  something like the following to
get leaf and hier node object classes which can go on the same structure, 
maybe this is helpful to you. 

type ('a, 'b) node = Leaf_node of 'a | Hier_node of 'b
type ('a,'b) instance =
    CellRef of (Atom.t * ('a, 'b) node * placement)
  | CellArrayRef of (Atom.t * ('a, 'b) node * placement * colrow * ipoint * ipoint)

class virtual leaf_intf  =
  object
    method virtual full_view : (leaf_intf, hier_intf) node
    ...
  end
and virtual hier_intf =
  object
    inherit leaf_intf
    method virtual leaf_view : (leaf_intf, hier_intf) node
    method virtual insts : (leaf_intf, hier_intf) instance list
  end

where instances will look like 

class some_leaf  =
  object (self)
    inherit leaf_intf
    method full_view = Leaf_node (self :> leaf_intf)
  end

class some_hier = 
  object (self)
    inherit hier_intf
    method full_view = Hier_node (self :> hier_intf)
    method leaf_view = Leaf_node (self :> leaf_intf)
   ...
  end

In all honesty though, when you need to do this kind of thing a lot you
should consider writing the code using algebraic data types rather than 
classes. I think that people used to a mostly OO language should avoid the 
OO features until they are comfortable with the classic ML way. 
 
-- Brian


-------------------
To unsubscribe, mail caml-list-request@inria.fr.  Archives: http://caml.inria.fr