gb2312与utf-8网页编码转换

特别感谢NETBOBO顶力帮助

本机测试运行环境：WindowsXP+CoreAmp

安装服务：nginx-0.7.64、mysql-5.1.42、php-cgi-5.3.1

有时候本机测试会有出入，应该以上传到服务器运行最终结果为准。这里给出的只是一种实现方法：

1。默认输出为UTF－8

2。如果目标页面编码为GB2312，转为UTF－8输出

3。本例中用到iconv转换，如果你的机器不支持则需要另行下载安装

4。代码：

 <?php $message="http://www.csdn.com";//http://www.163.com为GB2312,可作为测试用 $contents = @file_get_contents("$message"); function charsetstrv($url, $length=null){ if(is_null($length)) { $length = sizeof($url); } $init = 0xE6359A60; $a = 0x9E3779B9; $b = 0x9E3779B9; $c = 0xE6359A60; $k = 0; $len = $length; $mixo = array(); while( $len >= 12 ){ $a += ($url[$k+0] +box_32bit($url[$k+1]<<8) +box_32bit($url[$k+2]<<16) +box_32bit($url[$k+3]<<24)); $b += ($url[$k+4] +box_32bit($url[$k+5]<<8) +box_32bit($url[$k+6]<<16) +box_32bit($url[$k+7]<<24)); $c += ($url[$k+8] +box_32bit($url[$k+9]<<8) +box_32bit($url[$k+10]<<16) +box_32bit($url[$k+11]<<24)); $mixo = mix($a,$b,$c); $a = $mixo[0]; $b = $mixo[1]; $c = $mixo[2]; $k += 12; $len -= 12; } $c += $length; switch( $len ) { case 11: $c += box_32bit($url[$k+10]<<24); case 10: $c+=box_32bit($url[$k+9]<<16); case 9 : $c+=box_32bit($url[$k+8]<<8); case 8 : $b+=box_32bit($url[$k+7]<<24); case 7 : $b+=box_32bit($url[$k+6]<<16); case 6 : $b+=box_32bit($url[$k+5]<<8); case 5 : $b+=box_32bit($url[$k+4]); case 4 : $a+=box_32bit($url[$k+3]<<24); case 3 : $a+=box_32bit($url[$k+2]<<16); case 2 : $a+=box_32bit($url[$k+1]<<8); case 1 : $a+=box_32bit($url[$k+0]); } $mixo = mix( $a, $b, $c ); $mixo[2] = box_32bit($mixo[2]); if( $mixo[2] < 0 ){ return ( hexdec('1'. implode('', array_fill(0,PHP_INT_SIZE*2,'0'))) + $mixo[2] ); } else{ return $mixo[2]; } } //用正则取得网页编码方式 preg_match_all("/=([a-z0-9_]+)/",$contents,$array); $url=$array[0][0]; $bodytag = str_replace("=", "", $url); //判断并转换 if($bodytag=="gb2312") { $contents=iconv("GB2312","UTF-8",$contents); } //取得title,meta,body值 preg_match_all("/<title(.*?)<//title>/is",$contents,$title); preg_match_all("/<meta(.*?)/>/is",$contents,$meta); preg_match_all("/<body(.*?)body>/is",$contents,$body); for($i=0;$i<count($meta[0]);$i++){ if(preg_match("/keywords/i",$meta[0][$i])){ preg_match_all("/content=/"(.*?)/"/is",$meta[0][$i],$keywords); } if(preg_match("/description/i",$meta[0][$i])){ preg_match_all("/content=/"(.*?)/"/is",$meta[0][$i],$description); } } ?> <div align="center"> <table width="668" border="1"> <tr style="background:#C1DEF9; " mce_style="background:#C1DEF9; "> <th scope="col" width="228">Print</th> </tr> <tr> <td colspan="3" align="left"><?php echo 'title:'.strip_tags($title[0][0]); ?></td> </tr> <tr> <td colspan="3" align="left"><?php echo 'keywords:'.strip_tags($keywords[1][0]); ?></td> </tr> <tr> <td colspan="3" align="left"><?php echo 'description:'.strip_tags($description[1][0]); ?></td> </tr> <tr> <td colspan="3" align="left"><?php echo 'body:'.strip_tags($body[0][0]); ?></td> </tr> </table> </div>