PHP 감지 단어 필터링

/**
 * 금지어 필터링
 * 실행 효율: 글당 0.05초
 * @author liuxu
 *
*/
class Logic_BlackWord
{

 const APP_FORUM = 1;
 const APP_BLOG = 2 ;
 const APP_VOTE = 3;

 /**
 * 금지된 단어를 가져오기 위한 필터
 * @paramknown $txt
 * @return Ambigous
*/
 공개 함수 getHitList($txt)
 {
 $hitList = array();

 //对禁词分批过滤
 $max = $this->getMax();
 if($max)
 {
 $size = 1000;
 $ last = ceil($max/$size);
 for($page=1;$page<=$last;$page )
 {
 $result = $this->getHitListByPage($txt ,$page,$size);
 if($result) $hitList = array_merge($hitList,$result);
 }
 }

 $hitList2 = array();
 foreach($hitList as $hit=>$type)
 {
 $hitList2[$type][] = $hit;
 }

 return $hitList2;
 }

 비공개 함수 getMax()
 {
 $redis = Rds::factory();
 $memKey = 'blackWord_max';
 $max = $redis-> ;get($memKey);
 if($max===false)
 {
 $max = 0;
 $blackWord = new Model_BlackWord_BlackWord();
 $para['field '] = "MAX(id) AS max";
 $result = $blackWord->search($para);
 if(isset($result[0]['max'])) $max = $result[0]['max'];

 $redis->setex($memKey,300,$max);
 }

 $max 반환;
 }

 /**
 * 일괄 필터링하여 금지된 단어 얻기
 * @param 알 수 없는 $txt
 * @param number $page
 * @param number $size
 * @return multitype:Ambigous
*/
 비공개 함수 getHitListByPage($txt,$page=1,$size=1000)
 {
 $hitList = array();

 //분할批得到禁词树
 $wordTree = $this->getWordTreeByPage($page,$size);

 $txt = Strip_tags($txt);
 $txt = preg_replace('/[^a-zA-Z0-9\x{4e00}-\x{9fa5}]/iu','',$txt);

 $len = mb_strlen($txt,'UTF-8');
 for($i=0;$i<$len;$i )
 {
 $char = mb_substr($txt,$i,1 ,'UTF-8');
 if(isset($wordTree[$char]))
 {
 $result = $this->getHitListByTree(mb_substr($txt,$i,50, 'UTF-8'),$wordTree);
 if($result)
 {
 foreach($result as $hit=>$type)
 {
 $hitList[$ hit] = $type;
 }
 }
 }
 }

 return $hitList;
 }

 /**
 * 단어 금지 여부
 * @param str $txt
 * @param arr $wordTree
 * @return multitype:unknown
*/
 비공개 함수 getHitListByTree($txt,&$wordTree)
 {
 $len = mb_strlen($txt,'UTF-8');
 $point = & $wordTree;
 $ hit = '';
 $hitList = array();
 for($i=0;$i<$len;$i )
 {
 $char = mb_substr($txt,$ i,1,'UTF-8');
 if(isset($point[$char]))
 {
 $hit .= $char;
 $point = & $point[ $char];

 if(isset($point['type']))//匹配成功
 {
 $hitList[$hit] = $point['type'];
 }
 }
 else
 {
 break;
 }

 }

 return $hitList;
 }

 /**
 * 금지된 단어 트리를 일괄적으로 가져오기
 * @param int $page
 * @param int $size
 * @return arr:
*/
 비공개 함수 getWordTreeByPage($page=1,$size=1000)
 {
 $redis = Rds::factory();
 $memKey = 'blackWord_tree_'.$page.'_'.$size;
 $wordTree = $redis->get($memKey);
 if($wordTree===false)
 {
 $wordTree = array();
 $blackWord = new Model_BlackWord_BlackWord();
 $start = ($page-1)*$size;
 $end = $start $size;
 $para['where'] = "status=1 AND id>".$start." AND id<=".$end;
 $result = $blackWord->search($para);
 if($result)
 {
 foreach($result as $value)
 {
 if($value['word'])
 {
 $value[' word'] = preg_split('/(? $point = & $wordTree;
 foreach($value ['단어'] $char)
 {
 $point = & $point[$char];
 }

 $point['type'] = $value['type '];
 }
 }
 }

 $redis->setex($memKey,300,$wordTree);
 }

 return $wordTree ;
 }

}

PHP