C language input / output stream and file operation [II]

FILE Structure and buffer are discussed in depth

FILE Introduction to structure

---

Due to different compiler pairs FILE The implementation scheme of structure type is different Here only vs 2012 stdio.h Medium FILE Structure as an example ：

struct _iobuf{
    
	char* _ptr;// Point to the location information of the buffer to read and write data 
	int _cnt;//"W": Size of remaining bytes in buffer ,“r”: The byte size of the data write buffer 
	char* _base;// Point to the base address of the buffer 
	int _flag;// Status flag 
	int _file;// file type 
	int _charbuf;//
	int _bufsize;// Size of buffer 
	char* _tmpfname;// Temporary file name 
};
typedef struct _iobuf FILE;

---

buffer

To be completed ！！！

What's the difference between a text file and a binary file

---

The difference between text file and binary file

• text file Files are usually used to store content that is visible to the naked eye , for example ：.txt、.c、.dat Wait for the documents , We open these files through a text editor , You can understand the contents of the document .

• Binary Usually used to save videos 、 picture 、 Procedures, etc , Open these files with a text editor , You will see a lot of garbled code .

But the essence of physical storage is the same , It's all binary code , The reason why we can understand the content of text files , Because... Is used in the text file ASCII、UTF-8、GBK Equal character coding , The text editor can recognize these encoding formats , And convert the encoded value into characters to display .

Binary files use mp4、gif、exe And other special coding formats , Text editors don't recognize these encoding formats , It can only be parsed randomly according to the character encoding format , So it's a mess of characters , Some haven't even seen .

in general , Documents are the carrier of information , We need to know its storage type and interpret it in the correct form to get useful information .

---

fopen() Text mode and binary mode in (Windows Systems and Linux Line breaks for text files )

There is no essential difference between text files and binary files , Only the handling of newline symbols is different .

C Language program will \n As a line break , class UNIX/Linux The system will also \n As a line break , So the data in the program will be written to the text file intact , vice versa .

however Windows The system is different , It will \r\n As a newline character in a text file .

• stay Windows In the system , If you open a file as text , When reading a file , The program will all the in the file r\n Convert to one character \n. in other words , If two consecutive characters in a text file are \r\n, The program will discard the previous \r, Read in only \n.

• When writing to a file , Program meeting take \n convert to \r\n write in . in other words , If there are characters in the content to be written \n, Before writing this character , The program will automatically write a \r.

therefore , If the binary file is opened in text mode for reading and writing , The contents read and written may be different from the contents of the document .

In total , about Windows platform , For the sake of insurance , We'd better use "t" To open a text file , use "b" To open binary files . about Linux platform , Use "t" still "b" It doesn't matter , Since the default is "t", Then let's just write nothing .

---

C Language implementation getFileSize

#define _CRT_SECURE_NO_WARNINGS 
#include<stdio.h>
#include<stdlib.h>
#include<assert.h>

int main() {
    
	FILE* fp = fopen("test.txt", "r");
	fseek(fp, 0, SEEK_END);
	int len = ftell(fp);
	rewind(fp);
	fclose(fp);
	printf(" The length of the file is :%d", len);// In file  0D 0A
	return 0;
}

C Language implementation my_cp

stay C Language input / Output stream and file operation 【 One 】 The article gives my_cp But the efficiency of replication without consideration is low , Efficient file replication will be achieved here .

Realize file replication Main idea yes ： Open a buffer and constantly read data from the file into the buffer , Whenever the data is read once, the data in the buffer will be written to a new file at one time , Until the contents of the source file are all read .

Two key issues ：

1. How big a buffer should be opened ？
2. Buffer has no end flag , If the buffer is not full , How to determine the number of bytes written ？

• At present, most disk sectors are 4K The alignment of the , If the data written is not 4K An integer multiple of will be read across sectors , Reduce efficiency , So open up 4K The buffer .
• Each read returns the number of bytes read .

#include<stdio.h>
#include<stdlib.h>

int main(int arg ,char* argv[]){
    
    if(arg!=3){
    
        puts("input error");
        exit(0);
    }
    FILE * fpr = fopen(argv[1],"rb");
    if(fpr == NULL){
    
        puts("file open error");
        exit(0);
    }

    FILE * fpw = fopen(argv[2],"wb");
    if(fpw == NULL ){
    
        puts("file open error");
        exit(0);
    }

    int readSize = 1024 * 4;
    char* readData = (char*) malloc(readSize);
    if(readData == NULL){
    
        puts("malloc error");
        exit(0);
    }
    int realReadSize;
    while((realReadSize = fread(readData,1,readSize,fpr))>0){
    
        fwrite(readData,realReadSize,1,fpw);
    }
    free(readData);
    fclose(fpr);
    fclose(fpw);
    return 0;
}
// Linux Ubantu 20.4 
// gcc -o my_cp main.c
// ./my_cp sorceFileName newFileName

C Language implementation file insertion 、 Delete 、 change

To be completed ！！！