Path: utzoo!attcan!uunet!mcvax!enea!kth!draken!Urd!newsuser
From: newsuser@LTH.Se (LTH network news server)
Newsgroups: comp.lang.c++
Subject: Re: Friend specifier considered harmful
Message-ID: <1988Sep15.113208.21989@LTH.Se>
Date: 15 Sep 88 10:32:08 GMT
References: <61@cybaswan.UUCP> <1988Sep2.094800.16353@LTH.Se> <77@cybaswan.UUCP>
Reply-To: dag@Control.LTH.Se (Dag Bruck)
Organization: Dept. of Automatic Control, Lund Inst. of Technology, Sweden
Lines: 50

In article <77@cybaswan.UUCP> eeartym@cybaswan.UUCP (Dr R.Artym eleceng ) writes:
>In article <1988Sep2.094800.16353@LTH.Se> dag@Control.LTH.Se (Dag Bruck)
>writes:
>
>> I have seen many Modula-2 and Ada
>> programs where the writer has been "forced" to use very large modules
>> because of too strict encapsulation.
>
>Maybe this stemmed from lack of inheritance.  Why do you say strict
>encapsulation is to blame?

Yes, object-oriented programming may be the best way to get a
well-engineered program.  I still think that some ``escape'' mechanism
would improve Modula-2 and Ada programs, but probably not to the
degree where I will drop C++ (:-).

>> No, but I think my programs have improved by the use of a few friend
>> functions and classes.
>
>Can you supply any details?

The ``real'' work I do is too complicated to discuss in detail on the
net, but I will give an example in general terms:

In our application, my C++ program maintains a network of objects for
representing control system models.  As a typical example, the routine
for connecting two components of a block diagram needs access to some
internal information of so called terminals.  This is really just an
internal check to handle the special case of connecting to an
undefined enclosing environment (in the case of bottom-up
development).

>> Should we start a discussion of GOOD
>> design of classes?
>This would be a very worthwhile endeavour.

I will try to come up with a few topics in class design, but I'm sure
there are more competent and experienced people on the net.  I hope
there will be time for discussion at the USENIX C++ Conference in
Denver, CO.


Dag Bruck


-- 
Department of Automatic Control		Internet:  dag@control.lth.se
Lund Institute of Technology		UUCP:   ...!enea!control.lth.se!dag
P. O. Box 118
S-221 00 Lund, SWEDEN			Phone:	+46 46-108779