PHP センシティブワードフィルタリング

/**
 * 禁止ワードフィルタリング
 * 実行効率: 記事あたり 0.05 秒
 * @author liuxu
 *
*/
class Logic_BlackWord
{

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

 /**
 * 禁止単語を取得するためのフィルター
 * @param 不明 $txt
 * @return 曖昧な
*/
 public function getHitList($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;
 }

 プライベート関数 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);
 }

 return $max;
 }

 /**
 * バッチでフィルタリングして禁止単語を取得します
 * @param known $txt
 * @param number $page
 * @param number $size
 * @return multitype:Ambigous
*/
 private function getHitListByPage($txt,$page=1,$size=1000)
 {
 $hitList = array();

 //分批獲得禁止蠑
 $wordTree = $this->getWordTreeByPage($page, $size);

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

 $len = mb_strlen($txt,'UTF-8');
 for($i=0;$i {
 $char = mb_substr($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;
 }

 /**
 * 単語を禁止するかどうか
 * @param str $txt
 * @param arr $wordTree
 * @return multitype:unknown
*/
 プライベート関数 getHitListByTree($txt,&$wordTree)
 {
 $len = mb_strlen($txt,'UTF-8');
 $point = & $wordTree;
 $hit = '';
 $hitList = array();
 for($i=0;$i {
 $char = mb_substr($txt) ,$i,1,'UTF-8');
 if(isset($point[$char]))
 {
 $hit .= $char;
 $point = & $point[$char];

 if(isset($point['type']))//適合成功
 {
 $hitList[$hit] = $point['type'];
 }
 }
 else
 {
 Break;
 }

 }

 $hitList を返す;
 }

 /**
 * 禁止単語ツリーをバッチで取得します
 * @param int $page
 * @param int $size
 * @return arr:
*/
 プライベート関数 getWordTreeByPage($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['word'] as $char)
 {
 $point = & $point[$char];
 }

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

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

 return $wordTree;
 }

}

PHP