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=unavailable autolearn_force=no version=3.4.4
Path: 
 eternal-september.org!reader01.eternal-september.org!reader02.eternal-september.org!news.eternal-september.org!mx02.eternal-september.org!feeder.eternal-september.org!aioe.org!.POSTED!not-for-mail
From: "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de>
Newsgroups: comp.lang.ada
Subject: Re: Uninitialized out parameters.
Date: Fri, 8 Apr 2016 09:32:23 +0200
Organization: Aioe.org NNTP Server
Message-ID: <ne7mqt$1ute$1@gioia.aioe.org>
References: <3be79ab3-ebc7-4169-9713-d50349662403@googlegroups.com>
 <a99cbcf8-8a5c-41c4-a62b-519f0d3aec60@googlegroups.com>
 <e7ddc5a9-17d7-46f4-838a-994b61380b0b@googlegroups.com>
 <alpine.DEB.2.20.1604062240350.29458@debian> <ne3v4q$1od$1@loke.gir.dk>
 <alpine.DEB.2.20.1604071119210.30993@debian>
 <51041ba8-3243-46bc-b97d-b106d4b357e2@googlegroups.com>
 <ne6ouo$sm8$1@loke.gir.dk>
NNTP-Posting-Host: bqgfK7NL3xTHnr0WRaLl4g.user.gioia.aioe.org
Mime-Version: 1.0
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@aioe.org
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101
 Thunderbird/38.6.0
X-Notice: Filtered by postfilter v. 0.8.2
Xref: news.eternal-september.org comp.lang.ada:30048
Date: 2016-04-08T09:32:23+02:00
List-Id: <comp.lang.ada>

On 08/04/2016 01:02, Randy Brukardt wrote:
> "AdaMagica" <christ-usch.grein@t-online.de> wrote in message
> news:51041ba8-3243-46bc-b97d-b106d4b357e2@googlegroups.com...
>> Am Donnerstag, 7. April 2016 11:57:02 UTC+2 schrieb
>> Stefan...@uni-weimar.de:
>>
>>>> (2) The latter can be written in current Ada 2012: (that is, with TC1
>>>> implemented)
>>>>   function Y return T is (raise Program_Error);
>>>
>>> Great! I wasn't aware of that. I always thought, the expression in
>>> brackets had to be of type T.
>>>
>>> Stefan
>>
>> A raise_expression matches any type. See RM.
>
> Right. One can think of it sort of like "bottom" in a functional language -
> it represents the "not a value" of any type. (And touch it and you die.)
>
> OTOH, like me, you may not want to think of functional languages much. ;-)

Yes, better think it typed!

It is same as an IEEE float which values are real U {NaN, +Inf, -Inf}. 
Propagating exceptions is a sort of value.

In

    function Y return T is (raise Program_Error);

we assume that T consists of values of T values and exceptions of all 
sorts. That is weakly typed, a bit. But it also a way how to spell Y's 
post-condition: "T or Program_Error raised" could be a type Y returns. 
That would make an implementation (raise Constraint_Error) illegal.

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