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=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,2045dce6c7815bab
X-Google-Attributes: gid103376,public
From: bobduff@world.std.com (Robert A Duff)
Subject: Re: access types and dispatching subprograms
Date: 1996/06/23
Message-ID: <DtGx9M.F1s@world.std.com>#1/1
X-Deja-AN: 161724429
references: <4qjqre$44lo@news-s01.ny.us.ibm.net>
organization: The World Public Access UNIX, Brookline, MA
newsgroups: comp.lang.ada
Date: 1996-06-23T00:00:00+00:00
List-Id: <comp.lang.ada>


In article <4qjqre$44lo@news-s01.ny.us.ibm.net>,  <kbrun@ibm.net> wrote:
>   procedure disp(p: T_p);

You want "p: access T" here.  Access parameters, which are of an
anonymous access type, allow dispatching.  A named access type, such as
T_p, does not.

>   procedure disp(p: T1_p);

And "p: access T1" here.

>   procedure do_dispatch is
>      classp: TC_p := new T1;
>   begin
>      disp(classp);	-- this call won't dispatch, instead it causes
>                        -- a compiler error. what's wrong?

With the above changes, this will dispatch.

The reason you get a compile-time error is because we tried to set up
the rules so that if you *think* you are calling a dispatching
operation, but you are not, you will get a compile-time error if you
call it in a dispatching way.  As opposed to doing the unexpected thing
at run time.  This case is one example of that.

- Bob