From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <bruno.barras@inria.fr>
X-Original-To: caml-list@yquem.inria.fr
Delivered-To: caml-list@yquem.inria.fr
Received: from mail3-relais-sop.national.inria.fr (mail3-relais-sop.national.inria.fr [192.134.164.104])
	by yquem.inria.fr (Postfix) with ESMTP id B9089BC57
	for <caml-list@yquem.inria.fr>; Thu,  2 Sep 2010 15:06:30 +0200 (CEST)
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: AnEBAI8+f0yBaB4OmWdsb2JhbACTV40pFQEBAQEBCAsKBxEivTCFOwSKGQ
X-IronPort-AV: E=Sophos;i="4.56,308,1280700000"; 
   d="scan'208";a="56567979"
Received: from mx-a.polytechnique.fr ([129.104.30.14])
  by mail3-smtp-sop.national.inria.fr with ESMTP; 02 Sep 2010 15:06:30 +0200
Received: from argos.lix.polytechnique.fr (argos.lix.polytechnique.fr [129.104.11.2])
	by mx-a.polytechnique.fr (tbp 5.3.2/2.0.7) with ESMTP id o82D6Twu019012
	for <caml-list@yquem.inria.fr>; Thu, 2 Sep 2010 15:06:29 +0200
Received: from [129.104.11.145] (ausone.polytechnique.fr [129.104.11.145])
	by argos.lix.polytechnique.fr (Postfix) with ESMTP id 2BF04AAC82
	for <caml-list@yquem.inria.fr>; Thu,  2 Sep 2010 15:06:29 +0200 (CEST)
Message-ID: <4C7FA161.1060705@inria.fr>
Date: Thu, 02 Sep 2010 15:06:41 +0200
From: Bruno Barras <bruno.barras@inria.fr>
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.11) Gecko/20100713 Thunderbird/3.0.6
MIME-Version: 1.0
To: caml-list@yquem.inria.fr
Subject: Re: [Caml-list] extending records with Obj.magic
References: <20100830204956.2D927BC5D@yquem.inria.fr> <231FBA54-C413-4696-AA70-216253EDF2DA@math.harvard.edu>
In-Reply-To: <231FBA54-C413-4696-AA70-216253EDF2DA@math.harvard.edu>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-Spam: no; 0.00; untyped:01 val:01 val:01 bool:01 structurally:01 pervasives:01 pervasives:01 hash:01 equality:01 wrote:01 extensible:01 caml-list:01 int:01 int:01 seems:03 

On 08/30/2010 11:04 PM, Nicolas Ojeda Bar wrote:
> Hello,
>
> I need extensible records, and the code below seems to
> work. Are there any pitfall that I should be aware of?
> Could this mess up the GC?
>
> # type t0 = { a : int };
> # type t1 = { a2 : int; b : int };
> # value x1 = { a2 = 3; b = 5 };
> # value x0 : t0 = Obj.magic x1;
> value x0 = { a = 3 }
> # value x0' : t1 = Obj.magic x0;
> value x0' = { a2 = 3; b = 5 }

You should also be aware that structural equality (=) is untyped, so you 
might get surprised by this:

type t0 = { a : int; };;
# type t1 = { a2 : int; b : int; }
type t1 = { a2 : int; b : int; };;
# let x = {a2=3; b=1};;
val x : t1 = {a2 = 3; b = 1}
# let y = {a2=3; b=2};;
val y : t1 = {a2 = 3; b = 2}
# let x0 : t0 = Obj.magic x;;
val x0 : t0 = {a = 3}
# let y0 : t0 = Obj.magic y;;
val y0 : t0 = {a = 3}
# x0 = y0;;
- : bool = false

x0 and y0 look structurally equal, but they're not!

The same oddity will show up with Pervasives.compare and 
Pervasives.hash, just to mention those two.

Bruno.