去掉数据库导出的乱码（GBK版）

上次转发的那个 去掉半个字符的  只适用于 GB2312  

 

GB2312 中  第一个 和  第二个  字节   都在0x80以上

 

但是GBK 中  第一个字节  范围在  0x81-0xFE  第二个字节  范围在  0x40-0x7e 和 0x80-0xfe

 

"祎"字的  编码  B574  

 

原来的代码就会被误删

 

下面是我update的

 

/******************************************************************************* * * Module: GBK trim * Author: Richard ZHAN * Update: Sky * Description: Eliminate half Chinese character followed by a non Chinese character in a plain data file * * Change Log * * Date Name Description................. * 03/20/2009 Richard ZHAN Start Program(GB2312) * 10/22/2010 Sky Update GBK * *******************************************************************************/ #include <stdio.h> #include <stdlib.h> #include <unistd.h> #include <fcntl.h> #include <sys/types.h> #include <sys/stat.h> #include <errno.h> #include <strings.h> #define LEN 4096 int main (int argc, char *argv[]) { int rfd, wfd, len1, len2, i; char hi, *p1, *p2, str1[LEN], str2[LEN]; unsigned char ascii_hi = '/x7F'; if (argc != 3) { usage (); exit (1); } if ((rfd = open (argv[1], O_RDONLY)) == -1) { printf ("Cannot open read file!/n"); exit (1); } else if ((wfd = open (argv[2], O_RDWR | O_CREAT, 0644)) == -1) { printf ("Cannot open write file!/n"); close (rfd); exit (1); } else { hi = '/x0'; while ((len1 = read (rfd, str1, LEN)) > 0) { len2 = 0; bzero (str2, LEN); p2 = str2; for (p1 = str1, i = 0; i < len1; p1++, i++) { if ((unsigned char) (*p1) > '/x7F') { hi = *p1; p1++; i++; if(((unsigned char) (*p1) > '/x3F' && (unsigned char) (*p1) < '/x7F') || ((unsigned char) (*p1) > '/x7F')) { *p2++ = hi; *p2++ = *p1; len2 += 2; } else { *p2++ = *p1; len2++; } } else { *p2++ = *p1; len2++; } } if (write (wfd, str2, len2) != len2) { perror ("Encounter write error/n"); close (rfd); close (wfd); exit (1); } } if (len1 < 0) { perror ("Encounter read error/n"); close (rfd); close (wfd); exit (1); } } close (rfd); close (wfd); exit (0); } usage () { fprintf (stderr, "Usage: trim infile outfile/n"); return 0; }