From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,FREEMAIL_FROM
	autolearn=ham autolearn_force=no version=3.4.4
X-Google-Thread: 103376,9ed2b1e061ff269d
X-Google-NewGroupId: yes
X-Google-Attributes: gida07f3367d7,domainid0,public,usenet
X-Google-Language: ENGLISH,ASCII
Path: 
 g2news1.google.com!news4.google.com!proxad.net!feeder1-2.proxad.net!usenet-fr.net!gegeweb.org!aioe.org!not-for-mail
From: =?iso-8859-15?Q?Yannick_Duch=EAne_=28Hibou57=29?=
 <yannick_duchene@yahoo.fr>
Newsgroups: comp.lang.ada
Subject: Re: Subtypes boundaries
Date: Tue, 15 Jun 2010 01:31:47 +0200
Organization: Ada At Home
Message-ID: <op.vebcy90dxmjfy8@garhos>
References: <87nhicFa1kU1@mid.individual.net>
 <87nijlFa1kU2@mid.individual.net>
 <2b13f46b-65f9-450a-bb66-68c6093791c4@s9g2000yqd.googlegroups.com>
NNTP-Posting-Host: PS0QnrgsJzqTWblCjdHWPQ.user.speranza.aioe.org
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-15; format=flowed; delsp=yes
Content-Transfer-Encoding: Quoted-Printable
X-Complaints-To: abuse@aioe.org
X-Notice: Filtered by postfilter v. 0.8.2
User-Agent: Opera Mail/10.53 (Win32)
Xref: g2news1.google.com comp.lang.ada:11733
Date: 2010-06-15T01:31:47+02:00
List-Id: <comp.lang.ada>

Le Mon, 14 Jun 2010 23:01:17 +0200, Gautier write-only  =

<gautier_niouzes@hotmail.com> a =E9crit:
> To my surprise, it looks legal Ada.
The base type of IDEFIX is ASTERIX, by definition.

[ARM 2005 3.5(5)] just says:
> For a subtype_indication containing a range_constraint, either directl=
y =

> or as part of some other scalar_constraint, the type of the range shal=
l
> resolve to that of the type determined by the subtype_mark of the
> subtype_indication.

Nowhere in 3.5 I could find something explicitly stating when the error =
 =

must be detected (compile time or runtime).

If the ranges was defined using non-literal, like values of type ASTERIX=
,  =

and if these was variables, then there would be no other way except a  =

runtime detection of an error.

What disturbed me, is that a literal here, match an ASTERIX while it is =
 =

statically out-of range ?

Needs investigation in the RM (or may be the reference lacks something  =

here ?). What is the RM part involved here ?

-- =

There is even better than a pragma Assert: a SPARK --# check.
--# check C and WhoKnowWhat and YouKnowWho;
--# assert Ada;
--  i.e. forget about previous premises which leads to conclusion
--  and start with new conclusion as premise.