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=0.4 required=5.0 tests=BAYES_00,FORGED_MUA_MOZILLA
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Thread: 103376,b78c363353551702
X-Google-NewGroupId: yes
X-Google-Attributes: gida07f3367d7,domainid0,public,usenet
X-Google-Language: ENGLISH,ASCII-7-bit
Received: by 10.68.219.170 with SMTP id pp10mr994847pbc.1.1340312151605;
        Thu, 21 Jun 2012 13:55:51 -0700 (PDT)
Path: 
 l9ni4463pbj.0!nntp.google.com!news2.google.com!news.glorb.com!news.netfront.net!not-for-mail
From: Jeffrey Carter <spam.jrcarter.not@spam.not.acm.org>
Newsgroups: comp.lang.ada
Subject: Re: about the new Ada 2012 pre/post conditions
Date: Thu, 21 Jun 2012 13:55:48 -0700
Organization: Also freenews.netfront.net; news.tornevall.net
Message-ID: <js01ol$76g$1@adenine.netfront.net>
References: <jrsjr7$uuo$1@speranza.aioe.org>
 <4f0d55a9-83e1-44fe-8943-0c73a34a594d@googlegroups.com>
 <js00mr$cuc$1@munin.nbi.dk>
NNTP-Posting-Host: 184.20.201.198
Mime-Version: 1.0
X-Trace: adenine.netfront.net 1340312151 7376 184.20.201.198 (21 Jun 2012
 20:55:51 GMT)
X-Complaints-To: news@netfront.net
NNTP-Posting-Date: Thu, 21 Jun 2012 20:55:51 +0000 (UTC)
User-Agent: Mozilla/5.0 (X11; Linux i686;
 rv:12.0) Gecko/20120430 Thunderbird/12.0.1
In-Reply-To: <js00mr$cuc$1@munin.nbi.dk>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Date: 2012-06-21T13:55:48-07:00
List-Id: <comp.lang.ada>

On 06/21/2012 01:37 PM, Randy Brukardt wrote:
>
> "Hard-coded checks" prevent the compiler from doing call-site optimizations
> and tools from doing much of anything. They should be avoided. The solution
> is the pragma I showed earlier:
>
>     pragma Assertion_Policy (Pre =>  Check,
>                                             Pre'Class =>  Check,
>                                             Static_Predicate =>  Check,
>                                             Dynamic_Predicate =>  Check);
>
> put in *every* reusable package spec. They still can suppress the checks by
> manually deleting the pragma, but it will render command line switches and
> IDE checkboxes ineffective. And if they do delete the pragma, they've
> intentionally shot themselves in the foot, and it is no longer your (the
> package maintainers) problem. (Unless of course they want to spend extra
> $$$.)

Interesting. I wasn't aware of checks that can't be overridden by compiler 
options, and will be surprised if most compilers don't include a way to override 
these as well.

-- 
Jeff Carter
"Death awaits you all, with nasty, big, pointy teeth!"
Monty Python & the Holy Grail
20

--- Posted via news://freenews.netfront.net/ - Complaints to news@netfront.net ---