php关键词替换算法

<?php
/*

关键词匹配类
$str = "是是是是是范德萨下一年，下一年谁谁谁水水水水的洒落开是是是是军";
$key = new KeyReplace($str,array("下一年1"=>'http://baidu.com',"下一年"=>'baidu.com'));
echo $key->getResultText();
echo $key->getRuntime();
*/

$str = <<<ST
　　做开眼角手术需要开刀。开眼角手术材料的韩式技术，根计开眼角手术方案。
ST;

$g=file_get_contents("t");
$kw=unserialize($g);

$nkw=array();

foreach ($kw as $k=>$g){

if(strpos($k,"开眼角")!==false)
    $nkw[$k]=str_replace("php?", "g", $g);

}

// $nkw=array("开眼角"=>"http://hurl1","开眼角手术"=>"http://hurl2");

$key = new KeyWordReplace($str,$nkw);
echo $key->getResultText();

/*

url 中的特殊符号 如 ? . {} 需要做转义处理 ，替换完成之后再替换回来

• 关键词匹配类
• $str = "是是是是是范德萨下一年，下一年谁谁谁水水水水的洒落开是是是是军";
• $key = new KeyReplace($str,array("下一年1"=>'http://baidu.com',"下一年"=>'baidu.com'));
• echo $key->getResultText();

• echo $key->getRuntime();
  */
  function cmp2($a, $b){

  $len_a=mb_strlen($a,"utf-8");
  $len_b=mb_strlen($b,"utf-8");

  if ($len_a == $len_b) return 0;
  return ($len_a< $len_b) ? 1 : -1;
  }

class KeyWordReplace
{
public $keys = array();
public $text = "";
private $runtime = 0;
public $url = true;
public $stopkeys = array();
public $all = false;
/**

• @access public
• @param string $text 指定被处理的文章
• @param array $keys 指定字典词组array(key=>url,...) url可以是数组，如果是数组将随机替换其中的一个
• @param array $stopkeys 指定停止词array(key,...) 这里面的词将不会被处理
• @param boolean $url true 表示替换成链接否则只替换

• @param boolean $all true 表示替换所有找到的词，否则只替换第一次
  */
  public function __construct($text='',$keys=array(),$url=true,$stopkeys=array(),$all=true) {
  $this->keys = $keys;
  $this->text = $text;
  $this->url = $url;
  $this->stopkeys = $stopkeys;
  $this->all = $all;
  }

  /**

• 获取处理好的文章
• @access public

• @return string text
  */
  public function getResultText() {
  $start = microtime(true);
  $keys = $this->hits_keys();

  $keys_tmp = array_keys($keys);
  
  usort($keys_tmp,"cmp2");
  
  foreach($keys_tmp as $key){
  
      if(is_array($keys[$key])){
          $url = $keys[$key][rand(0,count($keys[$key])-1)];
      }else
          $url = $keys[$key];
  
          $this->text = $this->r_s($this->text,$key,$url);
  }
  $this->runtime = microtime(true)-$start;
  
  return $this->text;

  }
  /**

• 获取处理时间
• @access public

• @return float
  */
  public function getRuntime() {

  return $this->runtime;

  }

  /**

• 设置关键词
• @access public

• @param array $keys array(key=>url,...)
  */
  public function setKeys($keys) {

  $this->keys = $keys;

  }
  /**

• 设置停止词
• @access public

• @param array $keys array(key,...)
  */
  public function setStopKeys($keys) {

  $this->stopkeys = $keys;

  }
  /**

• 设置文章
• @access public

• @param string $text
  */
  public function setText($text) {

  $this->text = $text;

  }

  /**

• 用来找到字符串里面命中的关键词
• @access public

• @return array $keys 返回匹配到的词array(key=>url,...)
  */
  public function hits_keys(){
  $ar = $this->keys;
  $ar = $ar?$ar:array();
  $result=array();
  $str = $this->text;
  foreach($ar as $k=>$url){
  $k = trim($k);
  if(!$k)
  continue;
  if(strpos($str,$k)!==false && !in_array($k,$this->stopkeys)){
  $result[$k] = $url;
  }
  }
  return $result?$result:array();
  }

  /**

• 用来找到字符串里面命中的停止词
• @access public

• @return array $keys 返回匹配到的词array(key,...)
  */
  public function hits_stop_keys(){
  $ar = $this->stopkeys;
  $ar = $ar?$ar:array();
  $result=array();
  $str = $this->text;
  foreach($ar as $k){
  $k = trim($k);
  if(!$k)
  continue;
  if(strpos($str,$k)!==false && in_array($k,$this->stopkeys)){
  $result[] = $k;
  }
  }
  return $result?$result:array();
  }

  /**

• 处理替换过程
• @access private
• @param string $text 被替换者
• @param string $key 关键词
• @param string $url 链接

• @return string $text 处理好的文章
  */
  private function r_s($text,$key,$url){

  $tmp_text = $text;
  
  $stop_keys = $this->hits_stop_keys();
  
  $stopkeys = $tags = $a = array();
  if(preg_match_all("#<a[^>]+>[^<]*</a[^>]*>#su",$tmp_text,$m)){
      $a=$m[0];
  
      foreach($m[0] as $k=>$z){
          $z = preg_replace("#\##s","\#",$z);
          $tmp_text = preg_replace('#'.$z.'#s',"[_a".$k."_]",$tmp_text,1);
      }
  };
  
  if(preg_match_all("#<[^>]+>#s",$tmp_text,$m)){
      $tags = $m[0];
      foreach($m[0] as $k=>$z){
          $z = preg_replace("#\##s","\#",$z);
          $tmp_text = preg_replace('#'.$z.'#s',"[_tag".$k."_]",$tmp_text,1);
      }
  }
  if(!empty($stop_keys)){
      if(preg_match_all("#".implode("|",$stop_keys)."#s",$tmp_text,$m)){
          $stopkeys = $m[0];
          foreach($m[0] as $k=>$z){
              $z = preg_replace("#\##s","\#",$z);
              $tmp_text = preg_replace('#'.$z.'#s',"[_s".$k."_]",$tmp_text,1);
          }
      }
  }
  $key1 = preg_replace("#([\#\(\)\[\]\*])#s","\\\\$1",$key);
  
  if($this->url)
      $tmp_text = preg_replace("#(?!\[_s|\[_a|\[_|\[_t|\[_ta|\[_tag)".$key1."(?!ag\d+_\]|g\d+_\]|\d+_\]|s\d+_\]|_\])#us",'<a href="'.$url.'">'.$key.'</a>',$tmp_text,$this->all?-1:1);
  else
      $tmp_text = preg_replace("#(?!\[_s|\[_a|\[_|\[_t|\[_ta|\[_tag)".$key1."(?!ag\d+_\]|g\d+_\]|\d+_\]|s\d+_\]|_\])#us",$url,$tmp_text,$this->all?-1:1);
  
          if(!empty($a)){
  
              foreach($a as $n=>$at){
  
                  $tmp_text = str_replace("[_a".$n."_]",$at,$tmp_text);
  
              }
  
          }
          if(!empty($tags)){
  
              foreach($tags as $n=>$at){
  
                  $tmp_text = str_replace("[_tag".$n."_]",$at,$tmp_text);
  
              }
  
          }
          if(!empty($stopkeys)){
  
              foreach($stopkeys as $n=>$at){
  
                  $tmp_text = str_replace("[_s".$n."_]",$at,$tmp_text);
  
              }
          }
          return $tmp_text;

  }
  }