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=-0.3 required=5.0 tests=BAYES_00,
	REPLYTO_WITHOUT_TO_CC autolearn=no autolearn_force=no version=3.4.4
X-Google-Thread: 103376,ea5071f634c2ea8b
X-Google-NewGroupId: yes
X-Google-Attributes: gida07f3367d7,domainid0,public,usenet
X-Google-Language: ENGLISH,ASCII
Received: by 10.68.59.229 with SMTP id c5mr10332308pbr.6.1321782957251;
        Sun, 20 Nov 2011 01:55:57 -0800 (PST)
Path: 
 h5ni13015pba.0!nntp.google.com!news2.google.com!goblin3!goblin1!goblin2!goblin.stu.neva.ru!aioe.org!.POSTED!not-for-mail
From: "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de>
Newsgroups: comp.lang.ada
Subject: Re: Generic-Package Elaboration Question / Possible GNAT Bug.
Date: Sun, 20 Nov 2011 10:55:54 +0100
Organization: cbb software GmbH
Message-ID: <fe29npj98k5s$.1vzwcva8q22hk.dlg@40tude.net>
References: 
 <7bf9bc32-850a-40c6-9ae2-5254fe220533@f29g2000yqa.googlegroups.com>
 <wcc4nxzu5gf.fsf@shell01.TheWorld.com>
 <4295dc09-43de-4557-a095-fc108359f27f@y42g2000yqh.googlegroups.com>
Reply-To: mailbox@dmitry-kazakov.de
NNTP-Posting-Host: IhkZcqosmjDYmkmRk7+6yg.user.speranza.aioe.org
Mime-Version: 1.0
X-Complaints-To: abuse@aioe.org
User-Agent: 40tude_Dialog/2.0.15.1
X-Notice: Filtered by postfilter v. 0.8.2
Xref: news2.google.com comp.lang.ada:14470
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
Date: 2011-11-20T10:55:54+01:00
List-Id: <comp.lang.ada>

On Sat, 19 Nov 2011 15:36:33 -0800 (PST), Shark8 wrote:

> On Nov 19, 4:12�pm, Robert A Duff <bobd...@shell01.TheWorld.com>
> wrote:

>> In general, the design of Ada is that if a generic body is legal,
>> then every possible instance is legal (so the compiler need not
>> check the instances). �The design of C++ templates is very different
>> in this regard!
> 
> *nod* -- I understand that Ada's generics are quite different than C+
> +'s templates. Some of the reasons I've come across make a whole lot
> more sense to me for the Ada model than the C++, this "every possible
> instance" restriction is one such case. Who would want a system where
> a 'general' feature only worked for *some* instances. {Meaning the
> general feature wasn't actually .}

I don't know if Ada keeps that promise for generic bodies, but it certainly
does not for the specifications. Generic contracts are too weak for that.
Example:

generic
   type T is tagged private;
package Legal is
   type S is new T with null record;
   not overriding procedure Foo (X : in out S);
end Legal;

This and the body would successfully compile ignoring the fact that there
might exist tagged types with a primitive operation named Foo.

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de