Dynamic arrays and pointer notation

Pointer is a data type, similar to int or char. The data here is a memory address. Pointers are qualified with the types of values expected to be located in the memory at the referred addresses. As other data types, pointers could be used as types for variables, function arguments, array elements etc.

So, int **array is a variable of type int ** which means a pointer to a value of type int * located somewhere in the memory. In turn, int * means a pointer to a value of type int located somewhere in the memory. Or, in short, int ** is a pointer to a pointer to an int.

Expression new int*[4] means allocate in the memory an object of type int *[4] and return a pointer to it. Type int *[4] means an array of four elements of type int * where int *, as we already know, means a point to int value.

So, in the initialization, the types of the left part (int **) and of the right part (int *[]) are different. However, C++ is able to automatically convert arrays into pointers in case the type of array element is the same as the type of the value referred by the pointer. In our case, array element type is int * and the same is the type referred by the pointer. When an array is converted to a pointer, the resulting pointer will refer to the first element of the array.

In this declaration

int** array= new int*[4];

array is a variable that is initialized by the address of the first element of a dynamically allocated array with the element type int *.

If you have an array declared like

T array[N];

where T is some type then a pointer to the first element of the array can be declared like

T *p = array;

If to take into account your array then just T is equal to int *.

Simply put, a pointer is a variable that can store the address of another variable instead of an int or a float or some other value like a "normal" variable.

Like any other variable, the pointer itself also has an address so you can have a pointer to pointer, i.e. a pointer that stores the address of another pointer.