标题：不用iconv函数实现UTF-8编码转换GB2312的PHP函数
出处：沧海一粟
时间：Wed, 11 Oct 2006 16:08:27 +0000
作者：jed
地址：http://www.dzhope.com/post/109/

内容：
如果使用 iconv() 函数转换编码就相比比较简单了，不过很多虚拟主机里并不支持这个组件，我在网上找半天，才找到一个gb2312转utf-8的方法，但不能逆向转换。 

这个函数如下： 

/*******************************
//GB转UTF-8编码
*******************************/
function gb2utf8($gbstr) {
 global $CODETABLE;
 if(trim($gbstr)=="") return $gbstr;
 if(empty($CODETABLE)){
  $filename = dirname(__FILE__)."/gb2312-utf8.table";
  $fp = fopen($filename,"r");
  while ($l = fgets($fp,15))
  { $CODETABLE[hexdec(substr($l, 0, 6))] = substr($l, 7, 6); }
  fclose($fp);
 }
 $ret = "";
 $utf8 = "";
 while ($gbstr) {
  if (ord(substr($gbstr, 0, 1)) > 127) {
   $thisW = substr($gbstr, 0, 2);
   $gbstr = substr($gbstr, 2, strlen($gbstr));
   $utf8 = "";
   @$utf8 = u2utf8(hexdec($CODETABLE[hexdec(bin2hex($thisW)) - 0x8080]));
   if($utf8!=""){
 &nbsp; &nbsp;for ($i = 0;$i < strlen($utf8);$i += 3)
 &nbsp; &nbsp; $ret .= chr(substr($utf8, $i, 3));
 &nbsp; }
 &nbsp;}
 &nbsp;else
 &nbsp;{
 &nbsp; $ret .= substr($gbstr, 0, 1);
 &nbsp; $gbstr = substr($gbstr, 1, strlen($gbstr));
 &nbsp;}
 }
 return $ret;
}
//Unicode转utf8
function u2utf8($c) {
 for ($i = 0;$i < count($c);$i++)
 &nbsp;$str = "";
 if ($c < 0x80) {
 &nbsp;$str .= $c;
 } else if ($c < 0x800) {
 &nbsp;$str .= (0xC0 &#124; $c >> 6);
 &nbsp;$str .= (0x80 &#124; $c & 0x3F);
 } else if ($c < 0x10000) {
 &nbsp;$str .= (0xE0 &#124; $c >> 12);
 &nbsp;$str .= (0x80 &#124; $c >> 6 & 0x3F);
 &nbsp;$str .= (0x80 &#124; $c & 0x3F);
 } else if ($c < 0x200000) {
 &nbsp;$str .= (0xF0 &#124; $c >> 18);
 &nbsp;$str .= (0x80 &#124; $c >> 12 & 0x3F);
 &nbsp;$str .= (0x80 &#124; $c >> 6 & 0x3F);
 &nbsp;$str .= (0x80 &#124; $c & 0x3F);
 }
 return $str;
} 

因为gb2312都是双字节的，因此转换为utf-8就相对比较简单，但反之有很麻烦了，我尝试了一下： 

这样 

function utf82gb($utfstr)
{
 global $UC2GBTABLE;
 $okstr = "";
 if(trim($utfstr)=="") return $utfstr;
 if(empty($UC2GBTABLE)){
 &nbsp;$filename = dirname(__FILE__)."/gb2312-utf8.table";
 &nbsp;$fp = fopen($filename,"r");
 &nbsp;while($l = fgets($fp,15))
 &nbsp;{ $UC2GBTABLE[hexdec(substr($l, 7, 6))] = hexdec(substr($l, 0, 6));}
 &nbsp;fclose($fp);
 }
 $ulen = strlen($utfstr);
 for($i=0;$i<$ulen;$i++)
 {
 &nbsp;if(ord($utfstr[$i])<0x81) $okstr .= $utfstr[$i];
 &nbsp;else
 &nbsp;{
 &nbsp; if($ulen>$i+2)
 &nbsp; {
 &nbsp; &nbsp;$utfc = substr($utfstr,$i,3);
 &nbsp; &nbsp;$c = "";
 &nbsp; &nbsp;@$c = dechex($UC2GBTABLE[utf82u_3($utfc)]+0x8080);
 &nbsp; &nbsp;if($c!=""){
 &nbsp; &nbsp; &nbsp; $okstr .= chr(hexdec($c[0].$c[1])).chr(hexdec($c[2].$c[3]));
 &nbsp; &nbsp;}
 &nbsp; }
 &nbsp; else
 &nbsp; { $okstr .= $utfstr[$i]; }
 &nbsp;}
 &nbsp;}
 &nbsp;$okstr = trim($okstr);
 &nbsp;return $okstr;
} 

function utf82u_3($c)
{
 &nbsp; &nbsp; &nbsp;$n = (ord($c[0]) & 0x1f) << 12;
 &nbsp; &nbsp; &nbsp;$n += (ord($c[1]) & 0x3f) << 6;
 &nbsp; &nbsp; &nbsp;$n += ord($c[2]) & 0x3f;
 &nbsp; &nbsp; &nbsp;return $n;
} 

按这种方法，大部份字符也算是能转换成功的了，不过总是有点不妥之处，我把程序改成这样子： 

function utf82gb($utfstr)
{
 global $UC2GBTABLE;
 $okstr = "";
 if(trim($utfstr)=="") return $utfstr;
 if(empty($UC2GBTABLE)){
 &nbsp;$filename = dirname(__FILE__)."/gb2312-utf8.table";
 &nbsp;$fp = fopen($filename,"r");
 &nbsp;while($l = fgets($fp,15))
 &nbsp;{ $UC2GBTABLE[hexdec(substr($l, 7, 6))] = hexdec(substr($l, 0, 6));}
 &nbsp;fclose($fp);
 }
 $okstr = "";
 $utfstr = urlencode($utfstr);
 $ulen = strlen($utfstr);
 for($i=0;$i<$ulen;$i++)
 {
 &nbsp;if($utfstr[$i]=="%")
 &nbsp;{
 &nbsp; if($ulen>$i+2){
 &nbsp; &nbsp;$hexnext = hexdec("0x".substr($utfstr,$i+1,2));
 &nbsp; &nbsp;if($hexnext<127){
 &nbsp; &nbsp; $okstr .= chr($hexnext);
 &nbsp; &nbsp; $i = $i+2;
 &nbsp; &nbsp;}
 &nbsp; &nbsp;else{
 &nbsp; &nbsp; if($ulen>=$i+9){
 &nbsp; &nbsp; &nbsp;$hexnext = substr($utfstr,$i+1,8);
 &nbsp; &nbsp; &nbsp;$c = "";
 &nbsp; &nbsp; &nbsp;@$c = dechex($UC2GBTABLE[url_utf2u($hexnext)]+0x8080);
 &nbsp; &nbsp; &nbsp;if($c!=""){
 &nbsp; &nbsp; &nbsp; &nbsp;$okstr .= chr(hexdec($c[0].$c[1])).chr(hexdec($c[2].$c[3]));
 &nbsp; &nbsp; &nbsp;}
 &nbsp; &nbsp; &nbsp;$i = $i+8;
 &nbsp; &nbsp; }
 &nbsp; &nbsp;}
 &nbsp; }
 &nbsp; else
 &nbsp; { $okstr .= $utfstr[$i]; }
 &nbsp;}
 &nbsp;else if($utfstr[$i]=="+")
 &nbsp; $okstr .= " ";
 &nbsp;else
 &nbsp; $okstr .= $utfstr[$i];
 }
 $okstr = trim($okstr);
 return $okstr;
}
//三字节的URL编码转成的utf8字符转为unicode编码
function url_utf2u($c)
{
 $utfc = "";
 $cs = split("%",$c);
 for($i=0;$i<count($cs);$i++){
 &nbsp;$utfc .= chr(hexdec("0x".$cs[$i]));
 }
 $n = (ord($utfc[0]) & 0x1f) << 12;
 &nbsp;$n += (ord($utfc[1]) & 0x3f) << 6;
 &nbsp;$n += ord($utfc[2]) & 0x3f;
 return $n;
} 

一测试，发现完全OK，而且速度居然比上一个方法要快，我真是搞不懂这是什么原因了 

谁要 gb2312-utf8.table 这个文件请加我的QQ 2500875 IT柏拉图 或与 1877000 泡泡 联系 



Generated by Bo-blog 2.1.1 Release