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 autolearn=ham
	autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,7be3870dd9c9518f
X-Google-Attributes: gid103376,public
X-Google-ArrivalTime: 2002-12-15 11:05:37 PST
Newsgroups: comp.lang.ada
Path: 
 archiver1.google.com!news1.google.com!newsfeed.stanford.edu!bloom-beacon.mit.edu!ra.nrl.navy.mil!dca6-feed2.news.algx.net!allegiance!nntp.abs.net!uunet!dca.uu.net!ash.uu.net!world!news
From: Robert A Duff <bobduff@shell01.TheWorld.com>
Subject: Re: CONSTRAINT_ERROR - why?
User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2
Sender: news@world.std.com (Mr Usenet Himself)
Message-ID: <wccfzszp0hl.fsf@shell01.TheWorld.com>
Date: Sun, 15 Dec 2002 19:04:38 GMT
Content-Type: text/plain; charset=us-ascii
References: <3DFB7841.F898C02@t-online.de>
 <3%KK9.53003$vM1.4249406@bgtnsc04-news.ops.worldnet.att.net>
 <3DFB8495.A655C512@t-online.de> <616gta.hk3.ln@beastie.ix.netcom.com>
 <3DFC878D.B6966C37@t-online.de>
NNTP-Posting-Host: shell01.theworld.com
Mime-Version: 1.0
Organization: The World Public Access UNIX, Brookline, MA
Xref: archiver1.google.com comp.lang.ada:31857
Date: 2002-12-15T19:04:38+00:00
List-Id: <comp.lang.ada>

Alfred Hilscher <Alfred.Hilscher@t-online.de> writes:

> Dennis Lee Bieber schrieb:
> > 
> > From the user guide:
> > 
> > -gnato
> >    Enables overflow checking for integer operations. This causes GNAT
> > to generate slower and larger executable programs by adding code to
> > check for overflow (resulting in raising Constraint_Error as required
> > by standard Ada semantics). These overflow checks correspond to
> > situations in which the true value of the result of an operation may be
> 
> "... the result of an operation ..." The result of the operation "/"
> _does_ fit into a BYTE.

The Constraint_Error comes from the implicit conversion from universal
integer to BYTE, which happens to the *operands* of "/".  It's got
nothing to do with the "/" itself.

It's a Range_Check, by the way, not an Overflow_Check.  I don't know if
the -gnato switch understands that.

The relevant RM paragraph is 4.6(28).  It is a mistake, in my opinion
(although I didn't realize it at the time).  All the other operations on
modular types wrap around -- clearly type conversion should work the
same way.  (Randy will say that *none* of the operations should wrap
around, and I could go along with that, too.  But have one "gotcha"
operation work differently from all the rest is poor language design.)

> > outside the base range of the result type. The following example shows
> > the distinction:

- Bob