PE> Paul, I noticed that you used a similar but different method

 PE> of doing object-oriented C.  How about telling us what the

 PE> advantages are of your style?  Maybe it's me that should

 PE> be changing, but I'd like to know why.  BFN.  Paul.



Paul,



Apart from a few minor differences such as function names, there's only one
real difference. In your header files you declare your class as a structure
and I declare mine as a pointer. The effect is that all of your data member
are public and all of mine are private. I feel that making things private
is the better approach as it discourages fiddling with the internals of the
class.



My technique has one problem though. Because I typedef the class as a
pointer to a void, and the type checking isn't strict enough to pick this
up, it means that you can pass the wrong type of variables to my routines.
The type checking in C++ is strict enough from memory, but if I was
programming in C++ then I wouldn't be using this technique anyway :-)



What I'm playing with now is a hybrid of our two styles. Basically it
consists of defining a struct for the class as you do, but including a
pointer to the private data as I currently do. The definition looks
something like this:



    struct xyz

        {

        void    *privateStuff;

        }



This allows data to be kept private and for the compiler to do type checking.



What it also allows is for public data to be included as well:



    struct xyz

        {

        void    *privateStuff;



        /* public stuff follows */



        int     x;

        char    y;

        float   z:

        }



This is the best solution I've come up with so far.





Paul



--- GoldED/2 2.42.G0214+