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,
	T_FILL_THIS_FORM_SHORT autolearn=ham autolearn_force=no version=3.4.4
X-Google-Thread: 103376,e01befd2b86cac20
X-Google-Attributes: gid103376,public
X-Google-Language: ENGLISH,ASCII-7-bit
Path: 
 g2news1.google.com!news2.google.com!fu-berlin.de!uni-berlin.de!not-for-mail
From: "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de>
Newsgroups: comp.lang.ada
Subject: Re: problems with classes
Date: Mon, 4 Oct 2004 10:33:19 +0200
Message-ID: <1gnik0tlsc3av.1i85f33sd3tkx$.dlg@40tude.net>
References: <cjhtc3$8ul$03$1@news.t-online.com>
 <1c2igmxqnybiz.50giqp9woi08$.dlg@40tude.net>
 <cjk8e1$7ok$03$1@news.t-online.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Trace: news.uni-berlin.de 50quDb/rSnoqMN1egD5MnwtkK0TWCzEQSDP3QGSuhjegWSQhg=
User-Agent: 40tude_Dialog/2.0.12.1
Xref: g2news1.google.com comp.lang.ada:4645
Date: 2004-10-04T10:33:19+02:00
List-Id: <comp.lang.ada>

On Fri, 1 Oct 2004 20:41:50 +0200, Rick Santa-Cruz wrote:

>>> If this would be C++, I think I would know the answer... so in C++
>>> "something like this" exists, too. It is called pre-definition of a type.
>>> That means that, for example if I need the name of a type before I wanna
>>> specify it, I can just write:
>>> class MyType; and specify it later. Is this the case in Ada too?
>>
>> Close, but not quite. In your example you define the public view of a type
>> and later define its implementation. What you meant is a forward type
>> declaration, it looks exactly like in C++:
> Yes forward declaration ;).
> 
> But if "type Child_Class is new Parent_Class with private;" is not a forward 
> declaration then what is it then, or what is the sense of this soruce-line?

To publish the contract. It is a declaration of the public interface of the
type Child_Class. It states:

1. There is a type Child_Class
2. It is a descendant of Parent_Class (maybe an indirect one)

The clients of the package can rely on this contract. Forward declaration
(incomplete type declaration) does not define any completed contract. So
the following is illegal:

package Foo is -- ILLEGAL
   type X; -- Incomplete declaration
private
   type X is range 1..20; -- Type completion, "full view"

It makes no sense for public clients. No public view of X is defined, but
they cannot look into private. Compare with:

package Foo is -- LEGAL
   type X is private; -- True contract, "public view"
private
   type X is range 1..20; -- Type completion, "full view"

Also illegal is:

package Foo is -- ILLEGAL
   type X; -- Promise to complete it here
   type X is private; -- No, I will not

You cannot complete X in the private part, because you have promised that X
will be completed in the public part. So either:

package Foo is -- LEGAL
   ...
private
   type X; -- Incomplete declaration
   type X is range 1..20; -- Type completion and full view

or

package Foo is -- LEGAL
   type X; -- Incomplete declaration
   type X is range 1..20; -- Type completion and full view
private
   ...

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