problem with generic pointers

hey guys, I have 2 programs I am trying to figure out and the generic void pointers are really throwing me off. I realize the values are being swapped in the function but I don't understand how a=4 and b=7 are being sent in and then there in an array. When the numbers are passed in are they being converted to chars? If so what is 4 becoming?

Code:

// Generic Pointers

 #include <iostream>
 #include <iomanip>
 using namespace std;

 void q(void *a, void *b, int n) {

     unsigned char *ca = (unsigned char *)a;
     unsigned char *cb = (unsigned char *)b;
     unsigned char c;

     for (int i = 0; i < n; i++) {
         c     = ca[i];
         ca[i] = cb[i];
         cb[i] = c;
     }
 }

 int main() {

     int a = 4, b = 7;
     float x = 2.5f, y = 3.5f;

     q(&a, &b, sizeof(int));
     q(&x, &y, sizeof(float));

     cout << setprecision(1) << fixed;
     cout << "(a,b) = (" << a << ',' << b << ')' << endl;
     cout << "(x,y) = (" << x << ',' << y << ')' << endl; 
 }

Basically, the idea is to swap the values byte by byte.

In practice, we are more likely to use std::swap rather than such a byte copy swap because a correct swap of objects of class type may involve more than just swapping byte by byte.

That's because no "array's" are being passed in. You are passing in the memory location of a and b, not their actual values. Two pointers are required by q, and each call to it in the function main provides it with an address of some data. Since q is supposed to be "generic" in the sense that it should work with different sized data types, we also need to let q know how many bytes a and b are valid for from their base address. q() itself steps through the data a byte at a time up to n swapping the data. The "array's" you speak of are a way to dereference a block of memory at a specified offset. So it can be seen as a sort of shorthand for: (base address pointed to by ca) + (i * sizeof(element of ca)). In this case the size of each element is a byte.

so for example when c=ca[i] is happening, it is passing in the memory location of the 4?. And at that point I just have to realize a swap is happening because I won't know the actual values of the memory location?

Originally Posted by Thegame16

so for example when c=ca[i] is happening, it is passing in the memory location of the 4?.

No, the passing in of the memory location happens at the function call, not that point.

Originally Posted by Thegame16

And at that point I just have to realize a swap is happening because I won't know the actual values of the memory location?

That two bytes are being swapped should be obvious as the swap implementation is canonical. What is perhaps less obvious to you is that the idea is to treat the object, be it an int or a float, as an array of bytes, in order to swap the objects byte by byte.