Thread: Pointers and Classes

Hello all,

I was reading a tutorial on C++ programming, and in one example a class, let's call it, MyExample is defined. In the main function the following statements were written:

Code:

MyExample *p=new MyExample[2];
.
.
. 
p->SomeFunc(3,4);// .... (1)
p[1].SomeFunc(1,7);// ... (2)

where SomeFuc(int a, int b) is a public member function of the class that takes two int arguments.

To my understanding, the pointer p points to the first object of two objects of type MyExample that occupy consecutive memory locations. On the other hand, P[0] and p[1] are the first and second objects, respectively. I guess that's why we used -> with p in (1) because it's a pointer, while we used . with p[1] in (2) because it's an object, right?


It also seems to me that the statement (1) passes the int arguments 3 and 4 to both objects p[0] and [p1]. Is this how it works?

Thanks

Originally Posted by CNewProg

To my understanding, the pointer p points to the first object of two objects of type MyExample that occupy consecutive memory locations. On the other hand, P[0] and p[1] are the first and second objects, respectively. I guess that's why we used -> with p in (1) because it's a pointer, while we used . with p[1] in (2) because it's an object, right?

Yes.

Originally Posted by CNewProg

It also seems to me that the statement (1) passes the int arguments 3 and 4 to both objects p[0] and [p1]. Is this how it works?

No. For a pointer p, p[0] is equivalent to *(p + 0) which is equivalent to *p. (*p).foo is equivalent to p->foo. Hence, p->foo is equivalent to p[0].foo. I recommend the p[0].foo syntax here since p effectively represents a dynamic array, and what you are trying to do is to invoke the member function on p[0]. If p pointed to a lone object, or if it were used to traverse a sequence while pointing to individual objects in the sequence, then p->foo would be fine.