OK

QueryList.class.php

@@ -7,42 +7,93 @@
  * @author 			Jaeger
  * @email 			734708094@qq.com
  * @link            http://git.oschina.net/jae/QueryList
- * @version         1.6.1     
+ * @version         2.0.0     
+ *
+ * @example 
+ *
+ //获取CSDN移动开发栏目下的文章列表标题
+$hj = QueryList::Query('http://mobile.csdn.net/',array("title"=>array('.unit h1','text')));
+print_r($hj->jsonArr);
+
+//获取CSDN文章页下面的文章标题和内容
+$url = 'http://www.csdn.net/article/2014-06-05/2820091-build-or-buy-a-mobile-game-backend';
+$reg = array(
+    'title'=>array('h1','text'),    //获取纯文本格式的标题                   
+    'summary'=>array('.summary','text','input strong'), //获取纯文本的文章摘要，但保留input和strong标签
+    'content'=>array('.news_content','html','div a')    //获取html格式的文章内容，但过滤掉div和a标签
+    );
+$rang = '.left';
+$hj = QueryList::Query($url,$reg,$rang,'curl');
+print_r($hj->jsonArr);
+
+//继续获取右边相关热门文章列表的标题以及链接地址
+$hj->setQuery(array('title'=>array('','text'),'url'=>array('a','href')),'#con_two_2 li');
+//输出json数据
+echo $hj->getJson();
  */
 require 'phpQuery/phpQuery.php';
 class QueryList
 {
-    private $pageURL;
     private $regArr;
     public $jsonArr;
     private $regRange;
     private $html;
     private $outputEncoding;
     private $htmlEncoding;
+    private static $ql;
     /**
-     * 构造函数
+     * 静态方法，访问入口
      * @param string $page            要抓取的网页URL地址(支持https);或者是html源代码
-     * @param array  $regArr         【选择器数组】说明：格式array("名称"=>array("选择器","类型"),.......),【类型】说明：值 "text" ,"html" ,"属性" 
+     * @param array  $regArr         【选择器数组】说明：格式array("名称"=>array("选择器","类型"[,"标签列表"]),.......),【类型】说明：值 "text" ,"html" ,"属性" ,【标签列表】:可选，当【类型】值为text时表示需要保留的HTML标签，为html时表示要过滤掉的HTML标签
      * @param string $regRange       【块选择器】：指 先按照规则 选出 几个大块 ，然后再分别再在块里面 进行相关的选择
      * @param string $getHtmlWay     【源码获取方式】指是通过curl抓取源码，还是通过file_get_contents抓取源码
      * @param string $outputEncoding【输出编码格式】指要以什么编码输出(UTF-8,GB2312,.....)，防止出现乱码,如果设置为 假值 则不改变原字符串编码
      */
-    public function __construct($page, $regArr, $regRange = '', $getHtmlWay = 'curl', $outputEncoding = false)
+    public static function Query($page, $regArr, $regRange = '', $getHtmlWay = 'curl', $outputEncoding = false)
     {
+        if(!(self::$ql instanceof self))
+        {
+            self::$ql = new self();
+        }
+        self::$ql->_query($page, $regArr, $regRange, $getHtmlWay, $outputEncoding);
+        return self::$ql;
+    }
+    /**
+     * 重新设置选择器
+     * @param array $regArr   选择器数组
+     * @param string $regRange 块选择器
+     */
+    public function setQuery($regArr, $regRange = '')
+    {
+        $this->jsonArr = array();
+        $this->regArr = $regArr;
+        $this->regRange = $regRange;
+        $this->_getList();
+    }
+    /**
+     * 得到JSON结构的结果
+     * @return string
+     */
+    public function getJSON()
+    {
+        return json_encode($this->jsonArr);
+    }
+    private function _query($page, $regArr, $regRange, $getHtmlWay, $outputEncoding)
+    {
+        $this->jsonArr = array();
         $this->outputEncoding = $outputEncoding;
         if ($this->_isURL($page)) {
-            $this->pageURL = $page;
             if ($getHtmlWay == 'curl') {
                 //为了能获取https://
                 $ch = curl_init();
-                curl_setopt($ch, CURLOPT_URL, $this->pageURL);
+                curl_setopt($ch, CURLOPT_URL, $page);
                 curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false);
                 curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, false);
                 curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
                 $this->html = curl_exec($ch);
                 curl_close($ch);
             } else {
-                $this->html = file_get_contents($this->pageURL);
+                $this->html = file_get_contents($page);
             }
         } else {
             $this->html = $page;
@@ -55,13 +106,6 @@ class QueryList
             $this->_getList();
         }
     }
-    public function setQuery($regArr, $regRange = '')
-    {
-        $this->jsonArr = array();
-        $this->regArr = $regArr;
-        $this->regRange = $regRange;
-        $this->_getList();
-    }
     private function _getList()
     {
         $hobj = phpQuery::newDocumentHTML($this->html);
@@ -70,13 +114,14 @@ class QueryList
             $i = 0;
             foreach ($robj as $item) {
                 while (list($key, $reg_value) = each($this->regArr)) {
+                    $tags = isset($reg_value[2])?$reg_value[2]:'';
                     $iobj = pq($item)->find($reg_value[0]);
                     switch ($reg_value[1]) {
                     case 'text':
-                        $this->jsonArr[$i][$key] = trim(pq($iobj)->text());
+                        $this->jsonArr[$i][$key] = $this->_allowTags(pq($iobj)->html(),$tags);
                         break;
                     case 'html':
-                        $this->jsonArr[$i][$key] = trim(pq($iobj)->html());
+                        $this->jsonArr[$i][$key] = $this->_stripTags(pq($iobj)->html(),$tags);
                         break;
                     default:
                         $this->jsonArr[$i][$key] = pq($iobj)->attr($reg_value[1]);
@@ -89,15 +134,16 @@ class QueryList
             }
         } else {
             while (list($key, $reg_value) = each($this->regArr)) {
+                $tags = isset($reg_value[2])?$reg_value[2]:'';
                 $lobj = pq($hobj)->find($reg_value[0]);
                 $i = 0;
                 foreach ($lobj as $item) {
                     switch ($reg_value[1]) {
                     case 'text':
-                        $this->jsonArr[$i++][$key] = trim(pq($item)->text());
+                        $this->jsonArr[$i++][$key] = $this->_allowTags(pq($item)->html(),$tags);
                         break;
                     case 'html':
-                        $this->jsonArr[$i++][$key] = trim(pq($item)->html());
+                        $this->jsonArr[$i++][$key] = $this->_stripTags(pq($item)->html(),$tags);
                         break;
                     default:
                         $this->jsonArr[$i++][$key] = pq($item)->attr($reg_value[1]);
@@ -111,10 +157,6 @@ class QueryList
             $this->jsonArr = $this->_arrayConvertEncoding($this->jsonArr, $this->outputEncoding, $this->htmlEncoding);
         }
     }
-    public function getJSON()
-    {
-        return json_encode($this->jsonArr);
-    }
     /**
      * 获取文件编码
      * @param $string
@@ -157,4 +199,36 @@ class QueryList
         }
         return false;
     }
+    /**
+     * 去除特定的html标签
+     * @param  string $html 
+     * @param  string $tags 多个标签名之间用空格隔开
+     * @return string       
+     */
+    private function _stripTags($html,$tags)
+    {
+        $tagsArr = preg_split("/\s+/",$tags,-1,PREG_SPLIT_NO_EMPTY);
+        $p = array();
+        foreach ($tagsArr as $tag) {  
+            $p[]="/(<(?:\/".$tag."|".$tag.")[^>]*>)/i";  
         }  
+        $html = preg_replace($p,"",trim($html));  
+        return $html;  
+    }
+    /**
+     * 保留特定的html标签
+     * @param  string $html 
+     * @param  string $tags 多个标签名之间用空格隔开
+     * @return string       
+     */
+    private function _allowTags($html,$tags)
+    {
+        $tagsArr = preg_split("/\s+/",$tags,-1,PREG_SPLIT_NO_EMPTY);
+        $allow = '';
+        foreach ($tagsArr as $tag) {
+            $allow .= "<$tag> ";
+        }
+        return strip_tags(trim($html),$allow);
+    }
+}
+

README.md

@@ -1 +1 @@
-
+

demo/Searcher.class.php

@@ -48,7 +48,7 @@ require '../QueryList.class.php';
                $reg_znum='/([\d,]+) result(s)?/';
                $getHtmlWay = 'curl';
            }
-           $searcherObj = new QueryList($url,$this->regArr,$this->regRange,$getHtmlWay,false);
+           $searcherObj = QueryList::Query($url,$this->regArr,$this->regRange,$getHtmlWay,false);
            for($i=0;$i<count($searcherObj->jsonArr);$i++)
            {
                if($this->searcher=='baidu')

demo/demo.php

@@ -6,7 +6,7 @@ $url = "http://www.oschina.net/code/list";
 $reg = array("title"=>array(".code_title a:eq(0)","text"),"url"=>array(".code_title a:eq(0)","href"),"author"=>array("img","title"));
 $rang = ".code_list li";
 //使用curl抓取源码并以GB2312编码格式输出
-$hj = new QueryList($url,$reg,$rang,'curl','GB2312');
+$hj = QueryList::Query($url,$reg,$rang,'curl','GB2312');
 $arr = $hj->jsonArr;
 echo "<pre>";
 print_r($arr);
@@ -21,7 +21,7 @@ echo $json . "<hr/>";
 //采OSC内容页内容
 $url = "http://www.oschina.net/code/snippet_186288_23816";
 $reg = array("title"=>array(".QTitle h1","text"),"con"=>array(".Content","html"));
-$hj = new QueryList($url,$reg);
+$hj = QueryList::Query($url,$reg);
 $arr = $hj->jsonArr;
 echo "<pre>";
 print_r($arr);