RE: what is object oriented programming

["Steve Richter" <srichter@xxxxxxxxxxxxx>]

-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx
[mailto:rpg400-l-bounces@xxxxxxxxxxxx]On Behalf Of Hans Boldt
Sent: Tuesday, May 06, 2003 1:37 PM
To: rpg400-l@xxxxxxxxxxxx
Subject: Re: what is object oriented programming


Steve Richter wrote:
> Do you agree that an object is a "thing" that is accessed only thru
exposed
> interfaces? That no access to the internal components of the object, by
> definition, is allowed?
>
> Now if your programming manager tells you that a certain data struct can
> only be accessed by using a published set of procedures, doesnt that then
> make the data struct an object?
>
> Fields in the struct can be removed, added, changed.  The struct, if you
> work on a V5R2 system, can be added as a member of a larger data struct.
All
> these changes will not break any code that uses the struct thru the
> published set of procs that you are told to use.
>
> is such a data struct just a buffer definition?

Hans replied:
>To put it simply, yes, a data structure is just a buffer definition.

Is a c++ class with public data members also just a buffer definition?  If
the members are all private, but a public member function allows direct
access to the data member ( returns a mutable reference ), is the
class/struct also just a buffer definition?  Sounds like you would say yes.

The point is that it is how the struct is used and the design of the procs
that access the struct, rather than what the compiler allows or does not
allow, that provides the real world definition of an object.

-Steve