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,b87849933931bc93
X-Google-Attributes: gid103376,public
From: mheaney@ni.net (Matthew Heaney)
Subject: Re: Multiple Type Interfaces
Date: 1997/01/28
Message-ID: <mheaney-ya023280002801972134070001@news.ni.net>#1/1
X-Deja-AN: 212910383
references: <JSA.97Jan16141937@alexandria> <6PbObMf-3RB@herold.franken.de>
 <32EE352A.167EB0E7@cis.ohio-state.edu> <32EE6EF2.2487@watson.ibm.com>
 <5cm4raINN6ru@snoopy.cis.ohio-state.edu>
content-type: text/plain; charset=ISO-8859-1
organization: Estormza Software
mime-version: 1.0
newsgroups: comp.lang.ada
Date: 1997-01-28T00:00:00+00:00
List-Id: <comp.lang.ada>


In article <5cm4raINN6ru@snoopy.cis.ohio-state.edu>,
dgibson@snoopy.cis.ohio-state.edu (david scott gibson) wrote:


>Okay, but then assume you want to extend this *package* by adding some
>new operations to T2 (but not T1).  If you extend T2, what happens to
>the operation which had T2'Class as a parameter.  My understanding is
>that it would not be present in the interface of the new package, thus
>the new package would not be a true extension of the original one.
>What I was talking about (as applied to Ada) is really package
>extension, not type extension.

Yes, but if you have an operation that takes T2'Class in the same package
as the declaration of type T2, then what you're saying is that that
operation is non-primitive, and you specifically *don't* want it to be
inherited by types that derive from T2.

I wouldn't bother trying to make the operation directly visible to clients
of the child package - let clients get it from the parent package, where
it's declared - but if you insist, then you could simply rename it in the
child (I think).

I don't think it's such a big deal that the operation isn't directly
visible from the child package, because clients of the child package have
to with the parent anyway, so they've got the non-primitive operation if
they really want it:

with P.C;
package body Q is
...
   P.Op (<object of T2'Class>)

vs.

   P.C.Op (<object of T2'Class>)

--------------------------------------------------------------------
Matthew Heaney
Software Development Consultant
<mailto:matthew_heaney@acm.org>
(818) 985-1271