关于中日文和UNICODE之间编码的转换(2008-7-25 15:05:00)
标签:JIS SHIFT-JIS gb2312 Unicode 转换
最近一段做一些关于文字编码方面的东西,常常涉及到各种编码字符之间的转换。主要是做中日文方面的,包括中文gb2312, 日文JIS, SHIFT-JIS,以及他们和Unnicode码之间的转换。
一 GBK <==> Unicode
unsigned short GBK2UNI(unsigned short usGBK)
{
unsigned char szEUC[2] = { usGBK >> 8, usGBK & 0xFF };
unsigned short usUNI;
MultiByteToWideChar(
936,
0,
(LPCSTR)szEUC,
2,
&usUNI,
1
);
return usUNI;
}
unsigned short UNI2GBK(unsigned short usUNI)
{
unsigned char szGBK[3]={0};
unsigned short wzUNI[2] = { usUNI, 0 };
unsigned short usGBK;
WideCharToMultiByte(
936,
0,
wzUNI,
2,
(LPSTR)szGBK,
2,
0,
0
);
usGBK = (szGBK[0] << 8) | szGBK[1];
return usGBK;
}
二 SHIFT-JIS <==> Unicode
unsigned short SJIS2UNI(unsigned short usSJIS)
{
unsigned char szEUC[2] = { usSJIS >> 8, usSJIS & 0xFF };
unsigned short usUNI;
MultiByteToWideChar(
932,
0,
(LPCSTR)szEUC,
2,
&usUNI,
1
);
return usUNI;
}
unsigned short UNI2SJIS(unsigned short usUNI)
{
unsigned char szSJIS[3] = { 0 };
unsigned short wzUNI[2] = { usUNI, 0 };
unsigned short usSJIS;
WideCharToMultiByte(
932,
0,
wzUNI,
2,
(LPSTR)szSJIS,
2,
0,
0
);
usSJIS = (szSJIS[0] << 8) | szSJIS[1];
return usSJIS;
}
三 JIS <=> Unicode
unsigned short JIS2UNI(unsigned short usJIS)
{
unsigned char szEUC[2] = { (usJIS | 0x8080) >> 8, (usJIS | 0x8080) & 0xFF };
unsigned short usUNI;
MultiByteToWideChar(
20932,
0,
(LPCSTR)szEUC,
2,
&usUNI,
1
);
return usUNI;
}
unsigned short UNI2JIS(unsigned short usUNI)
{
unsigned char szJIS[3] = { 0 };
unsigned short wzUNI[2] = { usUNI, 0 };
unsigned short usJIS;
WideCharToMultiByte(
20932,
0,
wzUNI,
2,
(LPSTR)szJIS,
2,
0,
0
);
usJIS = (szJIS[0] << 8) | szJIS[1];
return usJIS;
}
四 JIS <=> SHIFT-JIS
unsigned short SJIS2JIS( unsigned short sjis )
{
unsigned short ubyte, lbyte;
if (((sjis >= 0x8140) && (sjis <= 0x9ffc)) ||
((sjis >= 0xe040) && (sjis <= 0xeffc)) )
{
ubyte = sjis >> 8;
lbyte = sjis & 0x00ff;
if ( (lbyte <= 0x3f) || (lbyte == 0x7f) ||
(lbyte >= 0xfd) ) return 0;
if ( ubyte >= 0xe0 ) ubyte -= 0xc0; else ubyte -= 0x80;
ubyte = (ubyte << 1) + 0x1f;
if ( lbyte >= 0x9f )
{
ubyte++;
lbyte -= 0x7e;
} else {
if ( lbyte >= 0x80 ) lbyte--;
lbyte -= 0x1f;
}
return ( ubyte << 8 ) + lbyte;
} else {
return 0;
}
}
unsigned short JIS2SJIS( unsigned short jis )
{
unsigned short ubyte, lbyte;
ubyte = jis >> 8;
lbyte = jis & 0x00ff;
lbyte += 0x1f;
if ( lbyte >= 0x7f ) lbyte++;
if ( lbyte <= 0x3f ) return 0;
if ( (ubyte & 0x0001) == 0 )
{
lbyte = jis & 0x00ff;
lbyte += 0x7e;
ubyte--;
if ( lbyte > 0xfd ) return 0;
}
ubyte -= 0x1f;
ubyte = ubyte >> 1;
ubyte += 0x80;
if ( ubyte >= 0xa0 ) ubyte += 0x40;
if ( ((ubyte >= 0x81) && (ubyte <= 0x9f)) ||
((ubyte >= 0xe0) && (ubyte <= 0xef)) )
{
return (ubyte << 8) + lbyte;
} else {
return 0;
}
}
UNI2JIS这个函数好像不太好用,其他的都经过测试,没有问题的。现在我还不知道具体原因,现在我从 Unicode转到JIS是分两个步骤的,第一个步骤是先将Unicode转到SHIFT-JIS,然后由SHIFT-JIS转到JIS。如果那位朋友知 道什么原因,欢迎在我博客里指点指点。
好了,就这些,希望能给需要的朋友们带来一点方便。