[revisiting the classic C language] ~x,%c,%d,%x, etc. in C language, the role of the address character in C language, and the consortium in C language

1. c In language %x、%c、%d、%x And so on.

%a,%A 					 Read in a floating point value 　　
%c 						 Read in a character  　　
%d 						 Read in decimal integers  　　
%i 						 Read in decimal , octal , Hexadecimal integer  　　
%o 						 Read in octal integers  　　
%x,%X 					 Read in hexadecimal integers  　　
%s 						 Read in a string , In case of space 、 End with tab or newline . 　　
%f,%F,%e,%E,%g,%G 		 It's used to enter real numbers , You can enter it in decimal or exponential form . 　　
%p 						 Read in a pointer  　　
%u 						 Read in an unsigned decimal integer  　　
%n 						 The number of equivalent characters of the value read so far  　　
%[] 					 Scan character sets  　　
%% 						 read % Symbol 

2. c Language address character & The role of

C Language address character & There are two common places for .

1. Pointer variable assignment

int m = 214;

int *i = &m;

2. In the formal parameter of a function, it appears before the formal parameter variable

void func(int* &i)

{
    

}

3. understand C In language Consortium

3.1 What is a consortium ？

stay C In language , The definition of variable is the process of allocating storage space . General , Each variable has its own storage space , So can different data types be stored in the same memory space （ Not stored at the same time ） Well ？

The answer is yes , This can be achieved by using a consortium . A consortium is also called a consortium , stay C The key words defining a consortium in the language are union.

3.2 Definition

The general form of defining a union type is ：

union  Union name 
{
    
 Membership table 
};

The member table contains several members , The general form of membership is ： Type specifier Member name . The number of bytes it occupies is the same as the number of bytes occupied by the largest data type in the member .

3.2.1 give an example

Method 1 ： Create a template first , Redefining variables

//  Create a consortium template union perdata
union perdata
{
    
    int Class;
    char Office;
};
//  Use the consortium template to create two variables a, b
union perdata a,b;

here ,perdata Is the name of the consortium , The name was chosen by us at will , But try to give it a meaningful name . It is equivalent to a template , You can use this template to define variables a、b. Don't forget to define union.

Method 2 ： Create templates and variables at the same time

//  Create a consortium template union perdata Define two variables at the same time a、b
union perdata
{
    
    int Class;
    char Office;
}a,b;

This is similar to method one .

Method 3 ： Omit the consortium name

union
{
    
    int Class;
    char Office;
}a,b;

Relative to method one and method two , The consortium name is omitted here . Although more concise , But because there is no name , You can't use this union to define new variables later .

Method four ： Use typedef

//  Consortium template union perdata Rename as perdata_U
typedef union perdata
{
    
    int Class;
    char Office;
}perdata_U;
//  Use a new name perdata_U Create two variables a, b
perdata_U a,b;

3.3 Initialize the consortium

The initialization of the union is different from that of the structure , A union can only store one value . The consortium has three initialization methods ：

perdata_U a;
a.Class = 10;
perdata_U b = a;				/* 1、 Initialize a union to another union of the same type ; */
perdata_U c = {
    20};				/* 2、 Initialize the first member of the Union ; */
perdata_U d = {
    .Office = 30};   /* 3、 according to C99 standard , Use the specified initializer . */

3.4 application

Application 1 ： Judge whether it is big end or small end
Understanding of big end and small end , Please move to this article , Portal

#include<stdio.h>
#include <stdbool.h>
union Utest
{
    
    short num;
    char c;
};
int main(void)
{
    
    union Utest u;
    u.num = 0x5566;
    printf("%x\n",u.c);//66,66 It's low , When reading memory , From low address to high address , What is stored in the low address is the low order , So it's a small end 
    return 0;
}

Application 2 ： Separate high and low bytes
The operation of separating high and low bytes is often encountered in single chip microcomputer , For example, the timer interrupt reset operation is often performed

(65535-200)/256,

(65535-200)%256

This kind of operation , One division consumes four machine cycles , Taking remainder also requires a series of complex operations , If such operations need to be performed many times in a short time, it will undoubtedly bring a huge burden to the program . In fact, all we need for these operations is the separation of high and low bytes of data , In this way, we can easily reduce this part of the cost by using the consortium .

Code ：

union div
{
    
    int n;     // n Store the high and low byte data to be separated in 
    char a[2]; //  stay keil c An integer in takes up two bytes ,char Take up a byte , therefore n And array a Same number of bytes 
}test;
test.n = 65535-200; //  After this sentence, it's all ok 了 , Let's visit test In the array a To fetch high and low byte data 
TH1 = test.a[0];    // test.a[0] The high-order data is stored in 
TL1 = test.a[1];    // test.a[1] Stored in test.n Low order data 

Union data is aligned by address . Whether the high-order data or the low-order data depends on the size end mode of the platform ,51 It's the big end ,stm32 The default is small end , If there are other compilers, please test yourself . Division is achieved with only one subtraction instruction 、 The operation of taking remainder , It is especially useful for high frequency timing .

=========================================================================

reference

c Language

Consortium

Big and small end judgment