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=0.2 required=5.0 tests=BAYES_00,INVALID_MSGID,
	REPLYTO_WITHOUT_TO_CC autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,ac5c3bc59168d76
X-Google-Attributes: gid103376,public
From: ncohen@watson.ibm.com (Norman H. Cohen)
Subject: Re: Subprogram Renaming
Date: 1996/04/12
Message-ID: <4kmgha$139s@watnews1.watson.ibm.com>#1/1
X-Deja-AN: 147188275
distribution: world
references: <md5:FE4AB546A8392541EDC1E3FE12E3D8AF>
  <316AEA8D.7600@csehp3.mdc.com> <kdd95h53l3.fsf@dsd.camb>
 <4kgde6$q8t@watnews1.watson.ibm.com> <4kjhg2$139s@watnews1.watson.ibm.com>
 <316E12F2.37BC@ehs.ericsson.se>
organization: IBM T.J. Watson Research Center
reply-to: ncohen@watson.ibm.com
newsgroups: comp.lang.ada
Date: 1996-04-12T00:00:00+00:00
List-Id: <comp.lang.ada>

In article <316E12F2.37BC@ehs.ericsson.se>, Jonas Nygren
<ehsjony@ehs.ericsson.se> writes: 

|> >    package Integer_Dequeues is
|> >       type Dequeue_Type is private;
...
|>
|>        procedure Remove_All (Dequeue: in out Dequeue_Type);
|>
|> >    private
|> >       type Dequeue_Type is ...;
|> >    end Integer_Dequeues;
|> >
|> >    with Integer_Dequeues;
|> >    package Integer_Stacks is
|> >       type Stack_Type is private;
...
|>
|>        procedure Remove_All (Stack: in out Stack_Type);
|>
|> >    private
|> >       use Integer_Dequeues;
|> >       type Stack_Type is new Dequeue_Type;
...
|>
|>        procedure Remove_All (Stack: in out Stack_Type)
|>           renames Remove_All; -- Does not work,
|>                               -- Gnat e.g. creates a recursive call (reported)
|>
|> >    end Integer_Stacks;

The explicit declaration of Remove_All in the visible part of
Integer_Stacks hides all homographs later in the same declaration, and
the version implicitly declared after the derived-type declaration is a
homograph of the explicitly declared version.  Thus you're trying to
rename the explicitly declared Remove_All as itself.

Here's an ugly solution: 

    package Integer_Stacks is
       type Stack_Type is private;
       ...

       procedure Remove_All (Stack: in out Stack_Type);

    private

       package Inner is
          type Stack_Parent_Type is new Integer_Dequeues.Dequeue_Type;
          procedure Parent_Remove_All
             (Stack: Stack_Parent_Type) renames Remove_All;
          ...
       end Inner;

       type Stack_Type is new Inner.Stack_Parent_Type;

       procedure Remove_All (Stack: in out Stack_Type)
          renames Parent_Remove_All;
          -- Inherited from Inner.Remove_All

       ...

    end Integer_Stacks;

At least the ugliness is hidden in the private part and has no runtime
overhead.

--
Norman H. Cohen    ncohen@watson.ibm.com