earlier today in chapter 3 I read about when he was going to do a system that returns a number within the size of the array, but with new every time.
the part of the chapter was called "Pseudo-Random Traversal of a set"

basically, the Mikey used prime numbers to make it possible. but why make the code so extremely advanced? here is what I think would give same results without having prime numbers, and mutch easier to make it a big size

I know how to put everything down in real code but I'm to lazy for it, hope you understands anyway

class theClass
{
int* array;
int currentsize;

}
Contructor(int size) // how big the size you will use will be
{
array = new int[size];
currentsize = size;
for ( i=0 ; i< size; ++i)
array = i;
}

int Next(bool restart)
{
int where = rand() % currentsize;
swap(array[where], array[currentsize-1]);
currentsize--;
return where;
}

of course you have to do all the checks also, but when you do it like this you get much less code, and much easier to understand also, and when you want to restart, you just change the currentsize. It doesnt matter that the numbers isnt in order since it is only rand you will use them for.

Feel free to comment

Ok, do you mean the override the swap function does or is it any more place?
If there is any other way I would gladly to know where it can occure.

And is it any problem if it's only doing it with the swap. I mean swap doesnt really take any time, and if it swap with itself no harm is done. because that just means it is the one that are going to be removed from the active list anyway