C language: Detailed explanation of secondary pointers (pointers to pointers)

Excerpt from: "Introduction to C Language" http://c.biancheng.net/c/

A pointer can point to a common type of data, such as int, double, char, etc., or it can point to a pointer type of data, such as int *, double *, char *, etc.

If one pointer points to another pointer, we call it 二级指针, or 指向指针的指针.

Suppose there is a variable a of type int, p1 is a pointer variable pointing to a, and p2 is a pointer variable pointing to p1. Their relationship is shown in the following figure:

Convert this relationship into C language code:

int a =100;
int *p1 = &a;
int **p2 = &p1;

A pointer variable is also a kind of variable, which also takes up storage space, and can also be used to &obtain its address. The C language does not limit the number of pointer levels. Each time a pointer is added, an asterisk must be added when defining a pointer variable *. p1 is a first-level pointer, which points to common types of data, and there is one when it is defined *; p2 is a second-level pointer, which points to a first-level pointer p1, and there are two when it is defined *.

If we want to define another three-level pointer p3, let it point to p2, then we can write:

int ***p3 = &p2;

The four-level pointer is similarly true:

int ****p4 = &p3;

In actual development, first-level pointers and second-level pointers are often used, and high-level pointers are hardly used.

When you want to get the data pointed to by the pointer, add a * to the first-level pointer, add two * to the second-level pointer, add three * to the third-level pointer, and so on, please see the code:

#include <stdio.h>
int main(){
    
    
    int a =100;
    int *p1 = &a;
    int **p2 = &p1;
    int ***p3 = &p2;
    printf("%d, %d, %d, %d\n", a, *p1, **p2, ***p3);
    printf("&p2 = %#X, p3 = %#X\n", &p2, p3);
    printf("&p1 = %#X, p2 = %#X, *p3 = %#X\n", &p1, p2, *p3);
    printf(" &a = %#X, p1 = %#X, *p2 = %#X, **p3 = %#X\n", &a, p1, *p2, **p3);
    return 0;
}

operation result:

100, 100, 100, 100
&p2 = 0X28FF3C, p3 = 0X28FF3C
&p1 = 0X28FF40, p2 = 0X28FF40, *p3 = 0X28FF40
&a = 0X28FF44, p1 = 0X28FF44, *p2 = 0X28FF44, **p3 = 0X28FF44

Take the three-level pointer p3 as an example to analyze the above code. ***p3Is equivalent to *(*(*p3)). *p3Get is p2of value, that p1address; *(*p3)get is p1of value, that aaddress; after three after the operation, "value" *(*(*p3))is derived avalue.

Assuming that the addresses of a, p1, p2, and p3 are 0X00A0, 0X1000, 0X2000, 0X3000, the relationship between them can be described by the following figure:

Inside the box is the value of the variable itself, and below the box is the address of the variable.