存档文件在c中提取

时间:2013-05-04 20:37:26

标签: c unix tar

我只是尝试在unix中编写类似tar命令的c代码。我的问题是提取存档文件。你会看到文件评论的近端“问题在这里”。我正在尝试提取档案文件。这是二进制文件。第一个字节是给我们文件的数量。第二个字节是第一个文件的名称长度。例如“file.c”长度为6.第三个字节是第一个文件的名称string.so“file.c”。并启动第二个文件的信息......转到最后一个文件的信息。所以最后开始每个文件包含之后和之后。

存档文件= N-I1-I2 ....- Ik-B1-B2 ....- Bk I = L-S-Z

N是文件数.1个字节 我是关于文件的信息。 B是文件包含。

L是文件名的长度。 1个字节 S是文件名字符串 Z是文件大小

所以,我可以从二进制文件中读取N,L,S但不能读取Z!我找不到它。

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/stat.h>
#include <sys/types.h>
#include <unistd.h>
#include <fcntl.h>
#define BUFSIZE 4096



struct stat st;

struct inputfiles{
  int numchar;
  char *filename;
  unsigned int filesize;
}file[255];

struct outputfiles{
  int num_char;
  char *file_name;
  unsigned int file_size;
}outfile[255];


int main(int argc,char *argv[])
{

  int copyfile(char *,char *);

  int i,j,k,m,h;
  int input_file,output_file;
  int fd;  
  int infile,outfile,arcfile;
  char buffer[BUFSIZE];
  char n;
  char flength;
 unsigned int file_len;
  char *f_name;

  char tempsize;
 unsigned  int sizeoffile;

  ssize_t nread,xread;




  input_file=argc-3;

  if(argc<=2)
    {
      fprintf(stderr,"Error: you must enter least 3 argument\n");
    exit(1);
    }
  else if(strcmp(argv[1],"-c")==0)
    {
      if((outfile=open(argv[2],O_WRONLY | O_CREAT,0644))==-1)
      {
      fprintf(stderr,"Error: Archive file can't open %s\n",argv[2]);
       remove(argv[2]);
      exit(1);
      }

      /*** write number of files into the archive file ***/
        write(outfile,(char*)&input_file,sizeof(char)); //UPDATED 


      j=0;    
      for(i=3;i<argc;i++)
    {
         file[j].numchar=strlen(argv[i]);


         /**** write filename size into archive file ****/ 
         write(outfile,(char*)&file[j].numchar,sizeof(char)); //UPDATED 



         file[j].filename=malloc(sizeof(file[j].numchar));
         strcpy(file[j].filename,argv[i]);

         /**** write filename into the archive file ****/
         write(outfile,file[j].filename,file[j].numchar);

         stat(argv[i],&st);
         file[j].filesize=st.st_size;     

         /**** write size of file into the archive file ****/
         write(outfile,&file[j].filesize,sizeof(int));               //UPDATED  HERE IS 4 BYTES

         j++;
    }   

      for(m=3;m<argc;m++)
    {
      if((infile=open(argv[m],O_RDONLY))==-1)
        {
          fprintf(stderr,"Error: File can't open %s\n",argv[m]);
          exit(1);
        }

      while((nread=read(infile,buffer,BUFSIZE))>0)
        {
          if(write(outfile,buffer,nread)<nread)
        {
          fprintf(stderr,"Error : input file size too much\n");
        }

          if(nread==-1)
        {
          fprintf(stderr,"Error occurred while reading.\n");
          exit(1);
        }
        }

    }
    }


/******* Extracting Archive File  *********/ 


  else if((strcmp(argv[1],"-x")==0))
    {
      if(argc!=3)
    {
      fprintf(stderr,"Error : you must enter 3 argument for extract \n");
      exit(1);
    }
      else
    {
     if((arcfile=open(argv[2],O_RDONLY))==-1)
       {
          fprintf(stderr,"Error:File can't open %s\n",argv[2]);
          remove(argv[2]);
          exit(1);
        }

     read(arcfile,&n,sizeof(char)); //read first byte of archive file
     output_file=(int)n;  // get number of output files
     printf("there is a %d files.\n",output_file);
     for(m=0;m<n;m++)   //loop for get information about each output file
       {
         read(arcfile,&flength,sizeof(char)); //read second byte
         file_len=((int)flength);   //get filename length


         f_name=malloc(file_len+1);   //malloc for filename
         read(arcfile,f_name,file_len);  //read size of filename length bytes and get filename string

         read(arcfile,&tempsize,sizeof(char)); //read size of file  <--- problem here
         sizeoffile=(int)tempsize;

     printf("file name length: %d\n",file_len);    
     printf("file name: %s\n",f_name);
     printf("file size: %d\n",sizeoffile);
       }

    }

    }
  else {
    fprintf(stderr,"invalid command line\n");
    exit(1);
  }
}

2 个答案:

答案 0 :(得分:1)

当你写出文件名和文件大小时,你将它们写成int s,占用4个字节:

write(outfile,&file[j].numchar,sizeof(file[j].numchar) /* = sizeof(int) */);
...
write(outfile,&file[j].filesize,sizeof(file[j].filesize) /* = sizeof(int) */);

但是当你把它们读出来时,你希望它们是char s,占用一个字节:

read(arcfile,&flength,sizeof(char));
...
read(arcfile,&tempsize,sizeof(char));

尝试将大小写为单个字节,或者更好,只需使用4个字节作为文件名和文件大小。 (更好的是因为你不必担心文件和文件名太长)


要将文件名和长度写为单个字节,请将它们转换为char s:

char to_write = file[j].numchar;
write(outfile, &to_write, sizeof(char));
...
to_write = file[j].filesize;
write(outfile, &to_write, sizeof(char));

答案 1 :(得分:0)

我明白了。问题是我的结构定义。我必须定义unsigned char numchar,而不是int numchar!。最后。谢谢你帮助我的朋友。