php中preg_match过滤URL非法参数程序

if(!preg_match("/^[a-z0-9_＼-]+$/i",$this->commentid))
    $this->_show_msg(L('illegal_parameters'));

$this->commentid = isset($_GET['commentid']) && trim(urldecode($_GET['commentid'])) ? trim(urldecode($_GET['commentid'])) : $this->_show_msg(L('illegal_parameters'));

function normalizeText($text, $length = null)
{
    $text = ＼Normalizer::normalize($text, ＼Normalizer::FORM_C);
    $text = preg_replace('/[^＼p{L}＼p{P}＼p{N}＼p{S}＼p{Zs}]/u', "", $text);
    $text = preg_replace('/^＼p{Z}*/u', "", $text);
    $text = preg_replace('/＼p{Z}*$/u', "", $text);
    if ($length !== null) {
        $text = mb_substr($text, 0, $length, 'utf-8');
    }
    return $text;
}

str="<a href=""a.htm"">a.htm</a><div>afefe</div>feaa<b>bbb</b> 

<a href=http://www.111cn.net id=""ggg"">ggg</a><abbr>测试</abbr>" 

Set re=new RegExp    

 re.IgnoreCase =true    

 re.Global=True    

 re.Pattern="(<(?!/?(a|p|b)(＼s|>))[^>]*>)"    

 str=re.replace(str," ")    

 response.Write str  

$str='<a href="a.htm">a.htm</a><div>afefe</div>feaa<b>bbb</b><a href=http://www.baidu.com id="ggg">ggg</a><abbr>测试</abbr>'; 
echo preg_replace('/(<(?!＼/?(a|p|b)(＼s|＼>))[^>]*>)/', '', $str);  
?> 

$str='<a href="a.htm">a.htm</a><div>afefe</div>feaa<b>bbb</b

/**
* @param $content
* @return mixed
* 过滤a标签保留内容
*/
public function delete_tags_a($content){
$content = preg_replace("#<a[^>]*>(.*?)</a>#is", "$1", $content);
return $content;
}

$array = (array)new SimpleXMLElement($xml_str);
$array = (array)simplexml_load_string($xml_str);
$array = json_decode(json_encode(simplexml_load_string($xml_str)),true); 都是返回 false

function findNum($str=''){
 if(empty($str)){return '';}
 $reg='/(＼d{4}(＼.＼d+)?)/is';//匹配数字的正则表达式
 preg_match_all($reg,$str,$result);
 if(is_array($result)&&!empty($result)&&!empty($result[1])&&!empty($result[1][0])){
  return $result[1][0];
 }
 return '';
}

$str=trim($str);
 if (preg_match('|(＼d+)|',$str,$r)) return $r[1];