UTF-8/UNICODE/简体中文/繁体中文之间的转换

Posted 2020-08-12

简介

这几天一直在研究中文的简体和繁体之间的转换问题，网上查了一下资料，在此进行整理和备份。

繁体中文有GBK码和BIG5码两种编码，简体中文一般使用的是GB2312编码。

这些编码之间的转换基本都是使用下列3个函数：LCMapString、WideCharToMultiByte和MultiByteToWideChar，其中还会牵涉到UNICODE码和UTF-8码这两种编码。

GB2312编码与GBK编码可以直接使用LCMapString转换，GB2312编码/GBK编码与BIG5编码则无法直接转换，必须使用UNICODE作为中间编码进行中转。

另外UTF-8编码是网络常用编码，如XML文件和网页基本都是使用这种编码，所以在此也一并研究了一下。

下面是我将GB2312/GBK/BIG5/UNICODE/UTF-8这5种编码之间的转换编写到一个函数的代码。

代码

  1 int Convert(void *sstr, int scp, void **dstr, int dcp)
  2 {
  3 #define CP_GBK        936
  4 #define CP_BIG5        950
  5 #define CP_UTF8        65001
  6 
  7     enum { _unicode, _utf8, _gb2312, _gbk, _big5 };
  8     enum { _wc2mb, _mb2wc, _sc2tc, _tc2sc };
  9 
 10     LCID lcid;
 11     void *src;
 12     void *dest;
 13     int cch;
 14     int scp0;
 15     int act;
 16     UINT cp;
 17 
 18     if (((scp < _unicode) || (scp > _big5)) ||
 19         ((dcp < _unicode) || (dcp > _big5)))
 20         return -1;
 21 
 22     src = NULL;
 23     dest = sstr;
 24     cch = 0;
 25     scp0 = scp;
 26 
 27     while (scp != dcp)
 28     {
 29         src = dest;
 30         switch (scp)
 31         {
 32         case _unicode:
 33             switch (dcp)
 34             {
 35             case _utf8:
 36                 scp = _utf8;
 37                 act = _wc2mb;
 38                 cp = CP_UTF8;
 39                 break;
 40             case _gb2312:
 41                 scp = ((scp0 == _big5) ? _gbk : _gb2312);
 42                 act = _wc2mb;
 43                 cp = CP_GBK;
 44                 break;
 45             case _gbk:
 46                 scp = _gbk;
 47                 act = _wc2mb;
 48                 cp = CP_GBK;
 49                 break;
 50             case _big5:
 51                 scp = _big5;
 52                 act = _wc2mb;
 53                 cp = CP_BIG5;
 54                 break;
 55             }
 56             break;
 57         case _utf8:
 58             switch (dcp)
 59             {
 60             case _unicode:
 61             case _gb2312:
 62             case _gbk:
 63             case _big5:
 64                 scp = _unicode;
 65                 act = _mb2wc;
 66                 cp = CP_UTF8;
 67                 break;
 68             }
 69             break;
 70         case _gb2312:
 71             switch (dcp)
 72             {
 73             case _unicode:
 74             case _utf8:
 75                 scp = _unicode;
 76                 act = _mb2wc;
 77                 cp = CP_GBK;
 78                 break;
 79             case _gbk:
 80             case _big5:
 81                 scp = _gbk;
 82                 act = _sc2tc;
 83                 break;
 84             }
 85             break;
 86         case _gbk:
 87             switch (dcp)
 88             {
 89             case _unicode:
 90             case _utf8:
 91             case _big5:
 92                 scp = _unicode;
 93                 act = _mb2wc;
 94                 cp = CP_GBK;
 95                 break;
 96             case _gb2312:
 97                 scp = _gb2312;
 98                 act = _tc2sc;
 99                 break;
100             }
101             break;
102         case _big5:
103             switch (dcp)
104             {
105             case _unicode:
106             case _utf8:
107             case _gb2312:
108             case _gbk:
109                 scp = _unicode;
110                 act = _mb2wc;
111                 cp = CP_BIG5;
112                 break;
113             }
114             break;
115         }
116 
117         switch (act)
118         {
119         case _wc2mb:
120             cch = WideCharToMultiByte(cp, 0, (wchar_t *)src, -1, NULL, 0, NULL, NULL);
121             dest = malloc(cch * sizeof(char));
122             WideCharToMultiByte(cp, 0, (wchar_t *)src, -1, (char *)dest, cch, NULL, NULL);
123             break;
124         case _mb2wc:
125             cch = MultiByteToWideChar(cp, 0, (char *)src, -1, NULL, 0);
126             dest = malloc(cch * sizeof(wchar_t));
127             MultiByteToWideChar(cp, 0, (char *)src, -1, (wchar_t *)dest, cch);
128             break;
129         case _sc2tc:
130             lcid = GetSystemDefaultLCID();
131             cch = LCMapString(lcid, LCMAP_TRADITIONAL_CHINESE, (char *)src, -1, NULL, 0);
132             dest = malloc(cch * sizeof(char));
133             LCMapString(lcid, LCMAP_TRADITIONAL_CHINESE, (char *)src, -1, (char *)dest, cch);
134             break;
135         case _tc2sc:
136             lcid = GetSystemDefaultLCID();
137             cch = LCMapString(lcid, LCMAP_SIMPLIFIED_CHINESE, (char *)src, -1, NULL, 0);
138             dest = malloc(cch * sizeof(char));
139             LCMapString(lcid, LCMAP_SIMPLIFIED_CHINESE, (char *)src, -1, (char *)dest, cch);
140             break;
141         }
142 
143         if (src && (src != sstr))
144         {
145             free(src);
146         }
147     }
148 
149     if (dstr)
150     {
151         *dstr = dest;
152     }
153     else
154     {
155         free(dest);
156     }
157 
158     return cch;
159 }

参数说明

sstr：[in]源字符串的首地址，由于可能是char *和wchar_t *两种数据类型，所以这里我设置为了void *类型 scp：[in]源字符串的编码方式，0：UNICODE编码、1：UTF-8编码、2：GB2312编码、3：GBK编码、4：BIG5编码 dstr：[out]目标字符串地址的指针，由于可能是char **和wchar_t **两种数据类型，所以这里我设置为了void **类型 dcp：[in]目标字符串的编码方式，取值范围与scp类似

函数使用

由于编码方式比较难记忆，所以我将任意两种编码的转化进行了如下定义

 #define UnicodeToUtf8(src, dest)      Convert((void *)(src), 0, (void **)(dest), 1)
#define UnicodeToGb2312(src, dest)    Convert((void *)(src), 0, (void **)(dest), 2)
#define UnicodeToGbk(src, dest)       Convert((void *)(src), 0, (void **)(dest), 3)
#define UnicodeToBig5(src, dest)      Convert((void *)(src), 0, (void **)(dest), 4)

#define Utf8ToUnicode(src, dest)      Convert((void *)(src), 1, (void **)(dest), 0)
#define Utf8ToGb2312(src, dest)       Convert((void *)(src), 1, (void **)(dest), 2)
#define Utf8ToGbk(src, dest)          Convert((void *)(src), 1, (void **)(dest), 3)
#define Utf8ToBig5(src, dest)         Convert((void *)(src), 1, (void **)(dest), 4)

#define Gb2312ToUnicode(src, dest)    Convert((void *)(src), 2, (void **)(dest), 0)
#define Gb2312ToUtf8(src, dest)       Convert((void *)(src), 2, (void **)(dest), 1)
#define Gb2312ToGbk(src, dest)        Convert((void *)(src), 2, (void **)(dest), 3)
#define Gb2312ToBig5(src, dest)       Convert((void *)(src), 2, (void **)(dest), 4)

#define GbkToUnicode(src, dest)       Convert((void *)(src), 3, (void **)(dest), 0)
#define GbkToUtf8(src, dest)          Convert((void *)(src), 3, (void **)(dest), 1)
#define GbkToGb2312(src, dest)        Convert((void *)(src), 3, (void **)(dest), 2)
#define GbkToBig5(src, dest)          Convert((void *)(src), 3, (void **)(dest), 4)

#define Big5ToUnicode(src, dest)      Convert((void *)(src), 4, (void **)(dest), 0)
#define Big5ToUtf8(src, dest)         Convert((void *)(src), 4, (void **)(dest), 1)
#define Big5ToGb2312(src, dest)       Convert((void *)(src), 4, (void **)(dest), 2)
#define Big5ToGbk(src, dest)          Convert((void *)(src), 4, (void **)(dest), 3)

测试代码如下：

 1 void main()
 2 {
 3     char *p0;
 4     char *p1;
 5 
 6     Gb2312ToBig5("中华人民共和国", &p0);
 7     printf("%s\n", p0);
 8     Big5ToGb2312(p0, &p1);
 9     printf("%s\n", p1);
10 
11     free(p0);
12     free(p1);
13 }