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=-0.8 required=5.0 tests=BAYES_00,INVALID_DATE
	autolearn=no autolearn_force=no version=3.4.4
Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site ucbvax.ARPA
Path: utzoo!watmath!clyde!burl!ulysses!ucbvax!usc-eclb.arpa!GBOOCH
From: GBOOCH@USC-ECLB.ARPA
Newsgroups: net.lang.ada
Subject: Re: design problem/?
Message-ID: <12142975372.32.GBOOCH@USC-ECLB.ARPA>
Date: Fri, 13-Sep-85 14:35:56 EDT
Article-I.D.: USC-ECLB.12142975372.32.GBOOCH
Posted: Fri Sep 13 14:35:56 1985
Date-Received: Sat, 14-Sep-85 08:16:02 EDT
References: <BRYAN@SU-SIERRA.ARPA>
Sender: daemon@ucbvax.ARPA
Organization: The ARPA Internet
List-Id: <comp.lang.ada>

For a heavilylayered system in which abstractions are built on
top of abstractions (and each layer hids its implementation), I tend to
use either access types (occasionally) or derived types (most often).

thus, in your example of the package file_stuff, I would probably write
the full type declaration for a_file as:

type a_file is new text_io.file_type;

This actually is a common issue (i.e. how tdo I get the implementation of
a private type to map excatly to another type). It would be
nice (sometimes) to have a fully type declaration in the private part
as a subtype, rather than a type, but Ada prevents this (and for
good reason, I believe...the alternative produces some strnage
problems).

egb
-------