From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <Christophe.Raffalli@univ-savoie.fr>
X-Spam-Checker-Version: SpamAssassin 3.1.3 (2006-06-01) on yquem.inria.fr
X-Spam-Level: 
X-Spam-Status: No, score=0.0 required=5.0 tests=none autolearn=disabled 
	version=3.1.3
X-Original-To: caml-list@yquem.inria.fr
Delivered-To: caml-list@yquem.inria.fr
Received: from discorde.inria.fr (discorde.inria.fr [192.93.2.38])
	by yquem.inria.fr (Postfix) with ESMTP id 2568ABC50
	for <caml-list@yquem.inria.fr>; Tue,  3 Oct 2006 21:59:06 +0200 (CEST)
Received: from smTp.neuf.fr (sp604003mt.neufgp.fr [84.96.92.56])
	by discorde.inria.fr (8.13.6/8.13.6) with ESMTP id k93Jx5IH008897
	for <caml-list@inria.fr>; Tue, 3 Oct 2006 21:59:05 +0200
Received: from [192.168.144.3] ([84.4.222.28]) by sp604003mt.gpm.neuf.ld
 (Sun Java System Messaging Server 6.2-5.05 (built Feb 16 2006))
 with ESMTP id <0J6K009MMSUHXR30@sp604003mt.gpm.neuf.ld> for
 caml-list@inria.fr; Tue, 03 Oct 2006 21:59:05 +0200 (CEST)
Date: Tue, 03 Oct 2006 21:59:45 +0200
From: Christophe Raffalli <Christophe.Raffalli@univ-savoie.fr>
Subject: Re: [Caml-list] float rounding
In-reply-to: <89957A27-4B6A-4FCF-A425-1468ECFA8B62@cmu.edu>
To: Sean McLaughlin <seanmcl@cmu.edu>
Cc: caml-list@inria.fr
Message-id: <4522C131.9010400@univ-savoie.fr>
MIME-version: 1.0
Content-type: text/plain; charset=ISO-8859-1; format=flowed
Content-transfer-encoding: 8BIT
References: <89957A27-4B6A-4FCF-A425-1468ECFA8B62@cmu.edu>
User-Agent: Thunderbird 1.5.0.7 (Macintosh/20060909)
X-Miltered: at discorde with ID 4522C109.000 by Joe's j-chkmail (http://j-chkmail.ensmp.fr)!
X-Spam: no; 0.00; christophe:01 raffalli:01 christophe:01 raffalli:01 univ-savoie:01 rounding:01 ocaml:01 ocaml:01 parser:01 compiler:01 compiler:01 printf:01 scanf:01 printf:01 sqrt:01 

Sean McLaughlin a écrit :
> Hello,
>
>   I'm using Ocaml for an interval arithmetic application.  I"m curious 
> about
> what the Ocaml parser/compiler does to float constants.  May I assume
> that for any constant I enter, eg. 3.1415... (for N digits of pi), that
> the compiler will give me a closest machine representable number?
> i.e., if I bound a constant by the previous and next floating point 
> value to
> that given me by the compiler,
> will it always be the case that my original (mathematical) constant 
> lies in that interval?
>

By the way, float constants need to be written in hexadecimal and this 
is missing to
the printf/scanf functions (it is what man printf says at least) ... 
just compute how many decimals you need to write the exact value of 
2^{-n} as a decimal float constant (0,5 0,25 0,125 0,625e-1 0,3125e-1 ...).

Then, I am not sure  the algorithm used to parse the constant will give 
you the correct value and you will never know how many decimals you need 
to write your constant.

The only problem with float and hexadecimal is to learn the hexadecimals 
of pi, e, ln 2, sqrt 2, ... ;-)

Christophe Raffalli