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 JAA22116; Wed, 31 Mar 2004 09:28:36 +0200 (MET DST)
X-Authentication-Warning: pauillac.inria.fr: majordomo set sender to owner-caml-list@pauillac.inria.fr using -f
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 JAA21636 for <caml-list@pauillac.inria.fr>; Wed, 31 Mar 2004 09:28:34 +0200 (MET DST)
Received: from out2.smtp.messagingengine.com (out2.smtp.messagingengine.com [66.111.4.26])
	by concorde.inria.fr (8.12.10/8.12.10) with ESMTP id i2V7SXYM000584
	for <caml-list@inria.fr>; Wed, 31 Mar 2004 09:28:34 +0200
X-Sasl-enc: 1F1HEVI0vz4BoU4co9v/5g 1080718111
Received: from [192.168.1.100] (unknown [218.81.120.251])
	by mail.messagingengine.com (Postfix) with ESMTP id D33F98DC742;
	Wed, 31 Mar 2004 02:28:30 -0500 (EST)
Date: Wed, 31 Mar 2004 15:28:13 +0800 (HKT)
From: Martin Jambon <martin_jambon@emailuser.net>
X-X-Sender: martin@localhost
To: Matt Gushee <mgushee@havenrock.com>
Cc: caml-list@inria.fr
Subject: Re: [Caml-list] examples of heterogenous collections (containers ?)
In-Reply-To: <20040331054113.GC19538@swordfish>
Message-ID: <Pine.LNX.4.40.0403311417570.2547-100000@localhost>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Miltered: at concorde by Joe's j-chkmail ("http://j-chkmail.ensmp.fr")!
X-Loop: caml-list@inria.fr
X-Spam: no; 0.00; caml-list:01 gushee:01 2004:99 objs:01 non-trivial:01 inherit:01 inherit:01 variants:01 variants:01 polymorphic:01 signatures:02 variant:02 variant:02 objects:02 objects:02 
Sender: owner-caml-list@pauillac.inria.fr
Precedence: bulk

On Tue, 30 Mar 2004, Matt Gushee wrote:

> On Tue, Mar 30, 2004 at 08:55:36PM -0800, briand@aracnet.com wrote:
> >
> > I'm embarking on that most typical of ewexamples, the heterogenous list.
> >
> > So I have objects:
> >
> > base_class
> >   virtual method
> >
> > derived_class_A
> >   inherit base_class
> >
> > derived_class_B
> >   inherit base_class
> >
> > ...
> >
> > And I would like to collect instances of derived_class_B,
> > derived_class_B etc.. into a data structure.
> >
> > The obvious example is that I'd like to collect the objects into a
> > list and do something like :
> >
> > List.iter
> >   (f obj ->
> >     obj#method)
> >   list_of_objs
>
> This will work if all the classes have identical signatures. Of course,
> that's not the case for most non-trivial examples.
>
> > or is (obj :> base_class)#method ?
>
> Probably not. First of all, you will need to cast the objects at the
> point where you put them into a list together ... thus, at the point
> where you do the method call, the cast has already been done.

And to keep trace of the original object, we would use polymorphic
variants:

class virtual ['a] base_class =
object
  method virtual variant : 'a
end

class ['a] derived_class_A =
object (self)
  inherit ['a] base_class
  method variant = `A self
end

and ['a] derived_class_B =
object (self)
  inherit ['a] base_class
  method variant = `B self
end

type variant = [ `A of variant derived_class_A
	       | `B of variant derived_class_B ]

let a = new derived_class_A
let b = new derived_class_B
let l = [ (a :> variant base_class);
	  (b :> variant base_class) ]
let list_of_variants = List.map (fun o -> o#variant) l


But maybe it is more convenient to not define a variant method, just
build a list of specialized objects:
  let list_of_variants = [ `A obj1; `B obj2; `C anything; ... ]

and then:
  let l = List.map (function `A o | `B o | `C o -> (o :> base_class))

(but here you loose the reference to the specialized version)


Martin

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners