[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Possible changes to abstract types proposal
- To: OOC Mailinglist <ooc-list@informatik.uni-kl.de>
- Subject: Re: Possible changes to abstract types proposal
- From: Tim Teulings <tim@edge.ping.de>
- Date: Mon, 1 Feb 1999 08:25:44 +0100
- Delivered-To: jnzimmer@donau.informatik.uni-kl.de
- In-Reply-To: <36B4F3EF.29C73F42@Informatik.Uni-Oldenburg.de>; from Marco Oetken on Mon, Feb 01, 1999 at 01:23:17AM +0100
- Mail-Followup-To: OOC Mailinglist <ooc-list@informatik.uni-kl.de>
- References: <3.0.1.32.19990106151246.0090e790@central.murdoch.edu.au> <36B4F3EF.29C73F42@Informatik.Uni-Oldenburg.de>
Hallo!
> What is the semantic fragile base class problem? I can't remember a description
> of this problem.
Classes that want to share some certain feature *must* inherit from the
same base class. This will give you problem, f.e. when you suddenly
want to put a class deeeply burried within your own application
class hierachy be put into a container. For a typical oberon-like
implementation of generic containers your class must now also inherit
from another node-class. This will leed to multiple inheritence wich is
not allowed on Oberon and which often is said to be the result of bad
design. Java solves this problem using interfaces (a very elegant
solution of the problem and a solution I would like to have in Oberon,
too), C++ uses templates.
> one feature? Why not changing OOC into a Component Pascal Compiler?
> I have read the documentations of the differences, recently, and I
> don't see any drawback, if like in BlackBox a Tool for Converting the
> sources is provided. And why not go further? I think that this language
> still can be improved.
Component Pascal is propertary, there is only one compiler
implementator. Also CP is not compatible to Oberon-2, which is in my
opinion the worst mistake the designers could make. If one extents O2
it must still by compatible in my opinion, normal programs still must run.
--
Gruß...
Tim.