吐血奉献:gb,unicode,utf8转码程序

<?
// Program by Donald,Milddragon Studio.
// Email: wilddragon#sina.com
// gb2312.txt请用google搜索下载

//初始化gb2312--unicode数组对应表作为全程变量，以提高处理速度
$____global_codetable=array();
$____global_filename=pathinfo($_SERVER["SCRIPT_FILENAME"]);
$____global_filename=$____global_filename["dirname"]."/gb2312.txt";
$____global_tmp=file($____global_filename);
while(list($key,$value)=each($____global_tmp))
{
if (strcmp($value{0},'#')!=0)
$____global_codetable[hexdec(substr($value,2,4))]=substr($value,9,4);
}
reset($____global_tmp);
while(list($key,$value)=each($____global_tmp))
{
if (strcmp($value{0},'#')!=0)
$____global_codetable2[hexdec(substr($value,9,4))]=hexdec(substr($value,2,4));
}
unset($____global_filename);
unset($____global_tmp);


/*
将带 㾏協格式的文本（可以包含其它ASCII字符）转换成gb2312格式的文本；
可以用于XML编码的转换
需要注意的是，函数不改变xml中关于编码的声明
*/
function unicode2gb($un)
{
if(!trim($un))
return $un;
$gb="";
global $____global_codetable2;
while(strlen($un)>0)
{
$p=strpos($un,"&#");
if ($p===FALSE)//串中已无unicode字符
{
$gb.=$un;
return $gb;
}
else
{
if ($p!=0)//串中unicode字符前缀不是第一个字符
{
$gb.=substr($un,0,$p);
$un=substr($un,$p);
}
$p=strpos($un,";");
if ($p===FALSE)//此前缀非unicode前缀,串中已无unicode字符
{
$gb.=$un;
return $gb;
}
else
{
$code=substr($un,2,$p-2);
$un=substr($un,$p+1);
if (strcasecmp($code{0},"x")==0)//unicode码16进制表示
{
$code=hexdec(substr($code,1));
}else
{
$code=intval($code);
}
$code=0x8080|$____global_codetable2[$code];
$gb.=chr((($code & 0xFF00)>>8) & 0xFF);
$gb.=chr($code & 0xFF);
}
}
}
return $gb;
}

/*
将 gb2312格式的文本（可以包含其它ASCII字符）转化为 带 㾏協格式的unicode文本；
可以用于XML编码的转换
需要注意的是，函数不改变xml中关于编码的声明
*/
function gb2unicode($gb)
{
if(!trim($gb))
return $gb;
$utf="";
global $____global_codetable;
while(strlen($gb)>0)
{
if (ord(substr($gb,0,1))>127)
{
$this=substr($gb,0,2);
$gb=substr($gb,2);
$code=$____global_codetable[hexdec(bin2hex($this))&0x7F7F];
$utf.="&#x".$code.";";
}
else
{
$utf.=substr($gb,0,1);
$gb=substr($gb,1);
}
}
return $utf;
}

/*
将utf8格式的文本转化为gb2312格式的文本;这与上述的unicode2gb不同,是二进制格式的转换
*/
function utf82gb($utf8)
{
if(!trim($utf8))
return $utf8;
global $____global_codetable2;
$gb="";
while(strlen($utf8)>0)
{
$c=substr($utf8,0,1);
$d=ord($c);
if (($d&0x80) == 0)//1位
{
$gb.=$c;
$utf8=substr($utf8,1);
}
else
if (($d&0xC0)==0x80)//错位
{
$utf8=substr($utf8,1);
}
else
if (($d&0xE0)==0xC0)//2位
{
$utf8=substr($utf8,2);
}
else
if (($d&0xF0)==0xE0)//3位
{
$d1=ord($utf8{1}) & 0x3F;
$d2=ord($utf8{2}) & 0x3F;
$d=$d & 0x0F;
$d=($d<<12) + ($d1 <<6) + $d2;
$code=0x8080|$____global_codetable2[$d];
$gb.=chr((($code & 0xFF00)>>8) & 0xFF);
$gb.=chr($code & 0xFF);
$utf8=substr($utf8,3);
}
else
if (($d&0xF8)==0xF0)//4位
{
$d1=ord($utf8{1}) & 0x3F;
$d2=ord($utf8{2}) & 0x3F;
$d3=ord($utf8{3}) & 0x3F;
$d=$d & 0x07;
$d=($d<<18) + ($d1 <<12) + ($d2 << 6) +$d3;
//$code=0x8080+getgb($d);
$code=0x8080|$____global_codetable2[$d];
$gb.=chr((($code & 0xFF00)>>8) & 0xFF);
$gb.=chr($code & 0xFF);
$utf8=substr($utf8,4);
}
else
{
$utf8=substr($utf8,1);
}
}
return $gb;
}

/*
将gb2312格式的文本转化为utf8格式的文本;这与上述的gb2unicode不同,是二进制格式的转换
*/
function gb2utf8($gb)
{
if(!trim($gb))
return $gb;
global $____global_codetable;
$utf8="";
while(strlen($gb)>0)
{
if (ord(substr($gb,0,1))>127)
{
$code=substr($gb,0,2);
$gb=substr($gb,2);
//echo "gb=$code;";
$code=bin2hex($code);
//echo "code=$code;";
$code=hexdec($code)&0x7F7F;
//echo "newcode=".dechex($code);
$code=$____global_codetable[$code];
//echo "unicode=$code";
$code=hexdec($code);
//11位：6+5
if (($code&0x7FF)==$code)
{
$utf8.=chr(0xC0|((($code&0x7C0)>>6)&0x3F));
$utf8.=chr(0x80|($code&0x3F));
}else
//16位：12+4
if (($code&0xFFFF)==$code)
{
$utf8.=chr(0xE0|((($code&0xF000)>>12)&0x3F));
$utf8.=chr(0x80|((($code&0xFC0)>>6)&0x3F));
$utf8.=chr(0x80|($code&0x3F));
//echo "16位==$utf8;\n";
}
else
//21位：18+3
if (($code&0x1FFFFF)==$code)
{
$utf8.=chr(0xF0|((($code&0x1C0000)>>18)&0x3F));
$utf8.=chr(0x80|((($code&0x3F000)>>12)&0x3F));
$utf8.=chr(0x80|((($code&0xFC0)>>6)&0x3F));
$utf8.=chr(0x80|($code&0x3F));
}
/*
else
//26位：24+2
if (($code&0x3FFFFFF)==$code)
{
$utf8.=chr(0xF8|((($code&0x3000000)>>24)&0x3F));
$utf8.=chr(0x80|((($code&0xFC0000)>>18)&0x3F));
$utf8.=chr(0x80|((($code&0x3F000)>>12)&0x3F));
$utf8.=chr(0x80|((($code&0xFC0)>>6)&0x3F));
$utf8.=chr(0x80|($code&0x3F));
}
else
//31位：30+1
if (($code&0x7FFFFFFF)==$code)
{
$utf8.=chr(0xFC|((($code&0x40000000)>>30)&0x3F));
$utf8.=chr(0x80|((($code&0x3F000000)>>24)&0x3F));
$utf8.=chr(0x80|((($code&0xFC0000)>>18)&0x3F));
$utf8.=chr(0x80|((($code&0x3F000)>>12)&0x3F));
$utf8.=chr(0x80|((($code&0xFC0)>>6)&0x3F));
$utf8.=chr(0x80|($code&0x3F));
}
//36位
else
{
//首字节全部作为前缀，无数据
$utf8.=chr(0x80|((($code&0xC0000000)>>30)&0x3F));
$utf8.=chr(0x80|((($code&0x3F000000)>>24)&0x3F));
$utf8.=chr(0x80|((($code&0xFC0000)>>18)&0x3F));
$utf8.=chr(0x80|((($code&0x3F000)>>12)&0x3F));
$utf8.=chr(0x80|((($code&0xFC0)>>6)&0x3F));
$utf8.=chr(0x80|($code&0x3F));
}
*/
}
else
{
$utf8.=substr($gb,0,1);
$gb=substr($gb,1);
}
}
return $utf8;
}
?>