文件遍历搜索（File System Traversal）

问题描述：

本文使用C语言实现了简单的文件遍历搜索功能并将搜索结果排序输出（测试平台为Ubuntu)，本程序接受以下格式输入：

输入

search.o [-n NAME] [-L] path

• path: 搜索开始的根地址，可以为相对或者绝对路径
• -n NAME: 被搜索文件的文件名模板，程序只输出文件名符合该模板的文件名，这个模板和shell风格的通配符一致：
  • *: 匹配任何文件，没有-n name参数时，默认仅有一个*号，即输出该路径下所有文件 
  • ?: 匹配任何单字符
  • [seq]: 匹配包含seq字符串的任何文件名
  • [!seq]: 匹配不包含seq字符串的任何文件名
• -L: 符号链接，如果有该参数，则符号链接也考虑在内，即程序会进入符号链接的文件下继续递归搜索

输出

• 一行输出一个搜索结果
• 在每一行，打印的结果以以下格式要求输出：
  • Mode: 一个九个字符的字符串，代表该文件的访问权限，实现的功能类似于命令： ls -l filename | cut -b 2-10 .
  • Size: the size of the file in a decimal number in bytes.一个十进制的数，代表该文件的大小（以字节计算）
  • Extended file name:完整的文件名，包含文件路径和文件名
• 输出结果按照文件名的ASCII码表排序输出

源码：search.c

//Written by Openking 2014-12-31

/*usage: search.o [-h] [-n NAME] [-L] path

positional arguments:
  path                  Path

optional arguments:
  -h, --help            show this help message and exit
  -n NAME, --name NAME  File name pattern
  -L, --followlinks     Follow symbolic links*/

#include <locale.h>
#include <fnmatch.h>
#include <stdio.h>
#include <errno.h>
#include <string.h>
#include <sys/types.h>
#include <dirent.h>
#include <stdbool.h>
#include <stdlib.h>
#include <sys/stat.h>

#define MAX_PATH_LENGTH	512

//global variables
char*name ="*";
bool Have_n = 0, Have_L = 0;
char dir_Path[10000][MAX_PATH_LENGTH];
char output[10000][MAX_PATH_LENGTH+20];
int num = 0;

void sort(char path1[][MAX_PATH_LENGTH],char path2[][MAX_PATH_LENGTH+20],int n)
{
    char temp1[MAX_PATH_LENGTH],temp2[MAX_PATH_LENGTH+20];
    int i,j,k;
    for(i = 0; i < n; i++)
    {
        k = i;
        for(j = i + 1; j < n; j++)
            if(strcmp(path1[k],path1[j]) > 0)
            	k = j;
        if(k != i)
        {
            strcpy(temp1,path1[i]);
            strcpy(temp2,path2[i]);
            strcpy(path1[i],path1[k]);
            strcpy(path2[i],path2[k]);
            strcpy(path1[k],temp1);
            strcpy(path2[k],temp2);
        }
    }
}

void doSearch(char *path)
{
	DIR *ptr_dir;
	struct dirent *dir_entry;
	int i = 0;
	char *child_path;
	char *file_path;

	child_path = (char*)malloc(sizeof(char)*MAX_PATH_LENGTH);
	memset(child_path, 0, sizeof(char)*MAX_PATH_LENGTH);

	file_path = (char*)malloc(sizeof(char)*MAX_PATH_LENGTH);
	memset(file_path, 0, sizeof(char)*MAX_PATH_LENGTH);

	ptr_dir = opendir(path);
	while((dir_entry = readdir(ptr_dir)) != NULL)
	{
		if(dir_entry->d_type == DT_DIR)    //directory
		{
			if(strcmp(dir_entry->d_name,".") == 0 ||
			   strcmp(dir_entry->d_name,"..") == 0)
				continue;

			sprintf(child_path, "%s/%s", path, dir_entry->d_name);
			//printf("[DIR]%s\n", child_path);
			doSearch(child_path);
			continue;
		}
		else if(dir_entry->d_type == DT_LNK)    //symbol link
		{
			sprintf(child_path, "%s/%s", path, dir_entry->d_name);
			//printf("[Link]%s\n", child_path);
			//printf("dir:%d lnk:%d reg:%d\n",DT_DIR,DT_LNK,DT_REG);
			struct stat sb;
			stat(child_path, &sb);
			if((sb.st_mode & S_IFMT) == S_IFDIR)
				if(Have_L)
					doSearch(child_path);
				else continue;
			else goto L1;
		}
		else if(dir_entry->d_type == DT_REG)
		{
			L1:
			if(fnmatch(name, dir_entry->d_name, FNM_PATHNAME|FNM_PERIOD) == 0)
			{
				sprintf(file_path, "%s/%s", path, dir_entry->d_name);
				//printf("[FILE]%s\n", file_path);
				sprintf(dir_Path[num], "%s",file_path);

				struct stat file;
				char power[] = "---------";
				int size_file = 0,mode = 0;
				lstat(file_path,&file);
				size_file = file.st_size;
				mode = file.st_mode;
				if(mode & S_IRUSR) power[0] = 'r';
				if(mode & S_IWUSR) power[1] = 'w';
				if(mode & S_IXUSR) power[2] = 'x';

				if(mode & S_IRGRP) power[3] = 'r';
				if(mode & S_IWGRP) power[4] = 'w';
				if(mode & S_IXGRP) power[5] = 'x';

				if(mode & S_IROTH) power[6] = 'r';
				if(mode & S_IWOTH) power[7] = 'w';
				if(mode & S_IXOTH) power[8] = 'x';
				sprintf(output[num++], "%s %d %s",power,size_file,file_path);
			}
		}
	}
	free(child_path);
	child_path = NULL;

	free(file_path);
	file_path = NULL;
}

int main(int argc, char*argv[], char *env[])
{
	int i,j;
	static char str1[] = "-n", str2[] = "-L";
	char*dir;

    	for(i = 1; i < argc; i++)    // Deal the strings
	{
		//printf("argv[%d]:%s\n", i, argv[i]);  //Output the command parameter
		if(strcmp(argv[i], str1) == 0)
		{
			Have_n = 1;
			name = argv[i+1];    //if have -n ,then record the name
			i++;      //skip the next name
			//printf("name: %s\n",name);
			continue;
		}
		else if(strcmp(argv[i], str2) == 0)
			Have_L = 1;
		else
		{
			dir = argv[i];     //record the directory
			int num_ch = strlen(dir);
			if(dir[num_ch - 1] =='/')
				dir[num_ch-1] = 0;
			//printf("directory: %s\n",dir);
		}
	}
	//printf("-L: %d; -n: %d\n",Have_L,Have_n);
	doSearch(dir);
	sort(dir_Path,output,num);
	for(i = 0; i < num; i++)
	{
		printf("%s\n",output[i]);
	}
	return 0;
}

测试输入：

search.o -n "*.py" -L ~/Desktop/workspace/python/

结果：

('rw-rw-r--', 1080, '/home/openking/Desktop/workspace/python/search.py')
('rw-rw-r--', 2442, '/home/openking/Desktop/workspace/python/shell.py')
('rw-rw-r--', 2262, '/home/openking/Desktop/workspace/python/test.py')