将各种编码格式的文本文件转换为ANSI格式

 

 

日常工作中，我们可能会遇到处理不同格式编码的文本文件的需求，这个问题如果处理不好，就会有中文乱码等棘手的问题。

以下这个文章写的很不错：

阮一峰：字符编码笔记：ASCII，Unicode和UTF-8 
 http://www.ruanyifeng.com/blog/2007/10/ascii_unicode_and_utf-8.html

理解这些编码知识后，我们就可以编写C++代码来完成任务了。下面的函数会把UTF-8, UNICODE Big-endian, UNICODE Little-endian这三种格式编码的文本文件转换成ANSI文本文件。参数为文本文件路径。只用于Windows平台。

  1 /*
  2  * 将路径fpath所指的文件从各种编码格式，转换为ANSI格式
  3  *
  4  * Copyright (c) 2013 赵子清， All rights reserved.
  5  *
  6  */
  7 
  8 #define CODE_FORMAT_ANSI        1
  9 #define CODE_FORMAT_UTF8        2
 10 #define CODE_FORMAT_UNICODE_LE  3
 11 #define CODE_FORMAT_UNICODE_BE  4
 12 
 13 typedef int ErrorCode;
 14 #define    ERR_OK                  0
 15 #define    ERR_FILE_OPEN_FAILED    1001
 16 
 17 #define SWAP16(x) \
 18     ((((x) & 0x00ff) << 8) | \
 19     (((x) & 0xff00) >> 8) )
 20 
 21 #define SWAP32(x) \
 22     ((((x) & 0x000000ff) << 24) | \
 23     (((x) & 0x0000ff00) <<  8) | \
 24     (((x) & 0x00ff0000) >>  8) | \
 25     (((x) & 0xff000000) >> 24) )
 26 
 27 #define SAFE_DELETE(x) if((x)!=0) {delete[] (x); (x) = 0; }
 28 
 29 ErrorCode  ConvertFormat(const char* fpath)
 30 {
 31 #ifdef _MSC_VER
 32     assert(fpath != 0);
 33 
 34     FILE* fp = ::fopen(fpath, "rb");
 35     if(fp == NULL)
 36         return ERR_FILE_OPEN_FAILED;
 37 
 38     int fmtFlag;
 39     int fmt = CODE_FORMAT_ANSI;
 40     ::fread(&fmtFlag, sizeof(int), 1, fp);
 41     ::fclose(fp);
 42 
 43     fmtFlag = SWAP32(fmtFlag);
 44     if((fmtFlag & 0xffffff00) == 0xefbbbf00)
 45         fmt = CODE_FORMAT_UTF8;
 46     else if((fmtFlag & 0xffff0000) == 0xfffe0000)
 47         fmt = CODE_FORMAT_UNICODE_LE;
 48     else if((fmtFlag & 0xffff0000) == 0xfeff0000)
 49         fmt = CODE_FORMAT_UNICODE_BE;
 50 
 51 
 52     if(fmt == CODE_FORMAT_ANSI)
 53         return ERR_OK;
 54 
 55     fp = ::fopen(fpath, "rb");
 56     char* txt = 0;
 57     wchar_t* wtxt = 0;
 58     long flen = 0L;
 59     ::fseek(fp, 0L, SEEK_END);
 60     flen = ftell(fp);
 61     ::rewind(fp);
 62     if(fmt == CODE_FORMAT_UTF8)
 63     {
 64         txt = new char[flen+1];
 65         ::fread(txt, 1, flen, fp);
 66         txt[flen] = '\0';
 67     }
 68     else if(fmt == CODE_FORMAT_UNICODE_LE || fmt == CODE_FORMAT_UNICODE_BE)
 69     {
 70         wtxt = new wchar_t[flen/2 +1];
 71         ::fread(wtxt, 2, flen/2, fp);
 72         if(fmt == CODE_FORMAT_UNICODE_BE)
 73         {
 74             for(int i=0; i < flen/2; i++)
 75                 wtxt[i] = SWAP16(wtxt[i]);
 76         }
 77         wtxt[flen/2] = L'\0';
 78     }
 79     
 80     ::fclose(fp);
 81 
 82     int nLen;
 83     wchar_t* pwstr = 0;
 84     char* pstr = 0;
 85     switch (fmt)
 86     {
 87     case CODE_FORMAT_UTF8:
 88         nLen = ::MultiByteToWideChar(CP_UTF8, 0, txt+3, -1, NULL, 0);
 89         pwstr = new wchar_t[nLen+1];
 90         nLen = ::MultiByteToWideChar(CP_UTF8, 0, txt+3, -1, pwstr, nLen);
 91         nLen = ::WideCharToMultiByte(CP_ACP, 0, pwstr, -1, NULL, 0, NULL, NULL);
 92         pstr = new char[nLen];
 93         ::memset(pstr, 0, nLen);
 94         nLen = ::WideCharToMultiByte(CP_ACP, 0, pwstr, -1, pstr, 
 95                                 nLen, NULL, NULL);
 96         break;
 97     case CODE_FORMAT_UNICODE_LE:
 98     case CODE_FORMAT_UNICODE_BE:
 99         nLen = ::WideCharToMultiByte(CP_ACP, 0, wtxt+1, -1, NULL, 0, NULL, NULL);
100         pstr = new char[nLen];
101         ::memset(pstr, 0, nLen);
102         nLen = ::WideCharToMultiByte(CP_ACP, 0, wtxt+1, -1, pstr, 
103                                         nLen, NULL, NULL);
104         break;
105     default:
106         break;
107     }
108 
109     fp = ::fopen(fpath, "wb");
110     ::fwrite(pstr, 1, nLen-1, fp);
111     ::fclose(fp);
112 
113     SAFE_DELETE(txt);
114     SAFE_DELETE(wtxt);
115     SAFE_DELETE(pstr);
116     SAFE_DELETE(pwstr);
117 
118 #endif
119 
120     return ERR_OK;
121 }

 

posted @ 2013-04-25 22:52  赵子清  阅读(5448)  评论(0)  编辑  收藏  举报