向东博客 专注WEB应用 构架之美 --- 构架之美，在于尽态极妍 | 应用之美，在于药到病除|赢在IT，Playin' with IT,Focus on Killer Application,Marketing Meets Technology.

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($utf8!=""){
   for ($i = 0;$i < strlen($utf8);$i += 3)
    $ret .= chr(substr($utf8, $i, 3));
  }
 }
 else
 {
  $ret .= substr($gbstr, 0, 1);
  $gbstr = substr($gbstr, 1, strlen($gbstr));
 }
}
return $ret;
}
//Unicode转utf8
function u2utf8($c) {
for ($i = 0;$i < count($c);$i++)
 $str = "";
if ($c < 0x80) {
 $str .= $c;
} else if ($c < 0x800) {
 $str .= (0xC0 | $c >> 6);
 $str .= (0x80 | $c & 0x3F);
} else if ($c < 0x10000) {
 $str .= (0xE0 | $c >> 12);
 $str .= (0x80 | $c >> 6 & 0x3F);
 $str .= (0x80 | $c & 0x3F);
} else if ($c < 0x200000) {
 $str .= (0xF0 | $c >> 18);
 $str .= (0x80 | $c >> 12 & 0x3F);
 $str .= (0x80 | $c >> 6 & 0x3F);
 $str .= (0x80 | $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)){
 $filename = dirname(__FILE__)."/gb2312-utf8.table";
 $fp = fopen($filename,"r");
 while($l = fgets($fp,15))
 { $UC2GBTABLE[hexdec(substr($l, 7, 6))] = hexdec(substr($l, 0, 6));}
 fclose($fp);
}
$ulen = strlen($utfstr);
for($i=0;$i<$ulen;$i++)
{
 if(ord($utfstr[$i])<0x81) $okstr .= $utfstr[$i];
 else
 {
  if($ulen>$i+2)
  {
   $utfc = substr($utfstr,$i,3);
   $c = "";
   @$c = dechex($UC2GBTABLE[utf82u_3($utfc)]+0x8080);
   if($c!=""){
      $okstr .= chr(hexdec($c[0].$c[1])).chr(hexdec($c[2].$c[3]));
   }
  }
  else
  { $okstr .= $utfstr[$i]; }
 }
 }
 $okstr = trim($okstr);
 return $okstr;
}

function utf82u_3($c)
{
     $n = (ord($c[0]) & 0x1f) << 12;
     $n += (ord($c[1]) & 0x3f) << 6;
     $n += ord($c[2]) & 0x3f;
     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)){
 $filename = dirname(__FILE__)."/gb2312-utf8.table";
 $fp = fopen($filename,"r");
 while($l = fgets($fp,15))
 { $UC2GBTABLE[hexdec(substr($l, 7, 6))] = hexdec(substr($l, 0, 6));}
 fclose($fp);
}
$okstr = "";
$utfstr = urlencode($utfstr);
$ulen = strlen($utfstr);
for($i=0;$i<$ulen;$i++)
{
 if($utfstr[$i]=="%")
 {
  if($ulen>$i+2){
   $hexnext = hexdec("0x".substr($utfstr,$i+1,2));
   if($hexnext<127){
    $okstr .= chr($hexnext);
    $i = $i+2;
   }
   else{
    if($ulen>=$i+9){
     $hexnext = substr($utfstr,$i+1,8);
     $c = "";
     @$c = dechex($UC2GBTABLE[url_utf2u($hexnext)]+0x8080);
     if($c!=""){
       $okstr .= chr(hexdec($c[0].$c[1])).chr(hexdec($c[2].$c[3]));
     }
     $i = $i+8;
    }
   }
  }
  else
  { $okstr .= $utfstr[$i]; }
 }
 else if($utfstr[$i]=="+")
  $okstr .= " ";
 else
  $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++){
 $utfc .= chr(hexdec("0x".$cs[$i]));
}
$n = (ord($utfc[0]) & 0x1f) << 12;
 $n += (ord($utfc[1]) & 0x3f) << 6;
 $n += ord($utfc[2]) & 0x3f;
return $n;
}

作者：jackxiang@向东博客 专注WEB应用 构架之美 --- 构架之美，在于尽态极妍 | 应用之美，在于药到病除
地址：http://jackxiang.com/post/857/
版权所有。转载时必须以链接形式注明作者和原始出处及本声明！