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=unavailable autolearn_force=no version=3.4.4
Path: 
 eternal-september.org!reader01.eternal-september.org!reader02.eternal-september.org!news.eternal-september.org!news.eternal-september.org!news.eternal-september.org!feeder.eternal-september.org!aioe.org!.POSTED!not-for-mail
From: Victor Porton <porton@narod.ru>
Newsgroups: comp.lang.ada
Subject: "NULL" returned by a function
Date: Sat, 09 Aug 2014 00:10:28 +0300
Organization: Aioe.org NNTP Server
Message-ID: <ls3ec1$v3q$1@speranza.aioe.org>
NNTP-Posting-Host: AnnUDmZwVERVUXyHDyOl5A.user.speranza.aioe.org
Mime-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 7Bit
X-Complaints-To: abuse@aioe.org
User-Agent: KNode/4.12.4
X-Notice: Filtered by postfilter v. 0.8.2
Xref: news.eternal-september.org comp.lang.ada:21584
Date: 2014-08-09T00:10:28+03:00
List-Id: <comp.lang.ada>

What if I want to make a function which returns either a value of type T or 
a value signifying "nothing" (like null pointer)?

To make things worse, T is a controlled type, so if we return just "access 
T" the finalization of the object of type T may be easily overlooked.

What is the best way to implement this? variant record? whatever?

Or maybe should I modify the type T to allow it contain "null" state? (In 
fact the particular T which I use can be easily modified to allow storing 
there null and checking whether it's state is null. But I think on the way 
not to be dependent on the ability to change the definition of T.)

-- 
Victor Porton - http://portonvictor.org