From: willi@twi.rhein-main.de (Thomas Wilhelmi)
To: mui@sunsite.Informatik.RWTH-Aachen.DE ("Kai Hofmann")
Date: Sat, 13 Apr 1996 15:22:15
Message-Id:
In-Reply-To: <60805488@informatik.uni-bremen.de>
X-Mailer: YAM 1.2 by Marcel Beck
Subject: Re: Interfaces wanted!
Resent-Message-Id: <"U6zsJ3.0.Nh1.EewRn"@sunsite>
Resent-From: mui@sunsite.Informatik.RWTH-Aachen.DE
Reply-To: mui@sunsite.Informatik.RWTH-Aachen.DE
X-Mailing-List: archive/latest/1028
X-Loop: mui@sunsite.informatik.rwth-aachen.de
Precedence: list
Resent-Sender: mui-request@sunsite.Informatik.RWTH-Aachen.DE
Content-Type: text
Content-Length: 1062
X-Lines: 29
Status: RO
Am 13-Apr-96 schrieb "Kai Hofmann":
> > A C++ Interface means that you provide a Class with all the Methods
> > as C++-Methods. And this means a lot more than some #define's.
> > You can't provide a C++-Interface for a MUI-public class without a
> > full C++-Interface for the basic MUI-System.
>
> Thats what I wanted :) - Is is easy to do (for you?) for this single class=
> ?
As I wrote above. It is really easy to write a C++-Class for
a single MUI-Class wich is derived from i.e. Area. But you need
a lot more. In a MUI C++-Class-library you need Con- and Destructors
wich Creates the classes and Objects (Constructors) and dispose
the objects and delete the classes (Destructors). And the destructors
have to decide wether they are allowed to dispose the objects and
to delete the classes.
I am working on such a class-library. But it is not ready now.
Ciao, Willi
------
Thomas Wilhelmi|D 61138 Niederdorfelden| willi@twi.rhein-main.de
|Taunusstrasse 14 | willi@sweet.commo.mcnet.de
|voice:+49 6101 531060 |
From: Jason S Birch
Message-Id: <199604160523.NAA11274@decadence>
Subject: Re: Interfaces wanted!
To: mui@sunsite.Informatik.RWTH-Aachen.DE
Date: Tue, 16 Apr 1996 13:23:46 +0800 (WST)
In-Reply-To: from "Thomas Wilhelmi" at Apr 16, 96 05:49:03 am
X-Mailer: ELM [version 2.4 PL25]
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Resent-Message-Id: <"yhhbk1.0.Av.UzoSn"@sunsite>
Resent-From: mui@sunsite.Informatik.RWTH-Aachen.DE
Reply-To: mui@sunsite.Informatik.RWTH-Aachen.DE
X-Mailing-List: archive/latest/1069
X-Loop: mui@sunsite.informatik.rwth-aachen.de
Precedence: list
Resent-Sender: mui-request@sunsite.Informatik.RWTH-Aachen.DE
X-Lines: 36
Status: RO
Content-Type: text/plain; charset="US-ASCII"
Content-Length: 1829
> > Is it only me, who does not know, how to C++ (can this be used as a verb ?)
> > on the Amiga ?? I'd really like to use C++ and MUI at the same time, but
> > after using Maxons C++ and GNU I switched back to Standard C. So my question
> > is: Is it worth doing a C++ interface for MUI???
>
> There is no big difference to program in C++ on the Amiga or on
> another computer. I would really say it is worth doing a C++ interface for
> MUI. Because if it exist the programming of a MUI-GUI ( :-) ) is
> much easier and quicker as without. (IMHO) But the first time you
> hve to program the C++-library itself is a lot of work.
Hmm... I'm not sure I can see how you can make a C++ interface to MUI
that uses C++'s inheritance (rather than BOOPSI's) without some
problems. Any custom class written in C++ (using C++'s inheritance and
method interface) would not be useable with another language, unlike a
custom class written to use BOOPSI (in any language).
I've seen a similar approach taken with Microsoft's Foundation
Classes. From what I've seen, they present an apparent C++ interface
to the outside world, but internally they use a mechanism very similar
to BOOPSI (ie. passing around ID's, etc) so you can still use the
classes without requiring C++. Any custom class that used C++'s
mechanisms, however, could only ever be used by someone using C++.
Still, I'll be curious to see the results. Perhaps I've completely
overlooked something. :-)
> Ciao, Willi
> Thomas Wilhelmi|D 61138 Niederdorfelden| willi@twi.rhein-main.de
--
Jason S Birch ,-_|\ email: jasonb@cs.uwa.edu.au