利用栈将html源码解析为节点树

最新推荐文章于 2021-06-24 13:07:24 发布

weixin_34258782

最新推荐文章于 2021-06-24 13:07:24 发布

阅读量263

点赞数

文章标签： php

原文链接：https://my.oschina.net/qidis/blog/535004

版权

为什么80%的码农都做不了架构师？>>>

<?php

/// 如何利用栈将html解析成节点树
/// 首先html是由一个个节点组成，最大的节点为<html></html>节点   她有两个子节点<head></head> 和<body></body>
/// 首先我们将<html>压入栈中再将<head>压入栈中遇到</head>出栈 <body>压入栈中遇到</body> 出栈最终<html>
/// 出栈可以看出出栈的一定是栈顶元素的子节点
/// 本程序假定html完全规范

/// todo 解决标签未闭合问题
/// todo 根据标签名访问
///       根据id访问
///       根据class访问
///      innerHtml 和 innerText

$tree_file = fopen("tree_file","w");
fclose($tree_file);

$siteurl="http://www.cnblogs.com/poissonnotes/archive/2010/05/28/1745996.html";
$str=get_content_by_url($siteurl);

$pattern="/<.*?>/";
$deep=0;

$global_false=true ;

$result=array();

$queue_id=0;

if(preg_match_all($pattern,$str,$matches))
{
   $count=count($matches[0]);
   $matches=$matches[0];
   $queue=array();
   $error = fopen("string_tagss","a");

   for($i=0;$i<$count;$i++)
   {
       $temptag = $matches[$i];

       $node["Id"]=$i;
       $node["ChildId"]=array();
       $node["html_tag"]=$temptag;
       /// 如果是单标签
       if(is_single_tag($temptag))
       {
           //echo "this is a single tag:   ";
           //echo $temptag."\n";
           $str_deep=make_string($deep);
           fwrite($error,$str_deep."\t".$temptag."\n");

           if(!empty($queue))
           {
               $queue[count($queue)-1]["ChildId"][]=$queue_id;
           }
           $queue_id++;
           $result[]=$node;
           continue ;
       }

       /// 如果是结束标签
       if(is_end_tag($temptag))
       {
           $temp_node = array_pop($queue);
           $queue[count($queue)-1]["ChildId"][]=$queue_id ; // 出队的节点一定是栈顶元素的子孩子
           $queue_id++;
           $result[]=$temp_node;
       }
       else
       {
           /// 如果是开始标签
           $str_deep=make_string($deep);
           fwrite($error,$str_deep."\t".$temptag."\n");
           $queue[] = $node;
           $deep++;
       }
   }

   if($global_false)
   {
       echo "ok\n" ;
   }
   else
   {
       echo "error\n" ;
   }

   //制作关系图谱不包含结束标签
   //<html>
   //   <head>
   //           ............
   //       ............
   //   <body>
   //              ............
        //              ............

   $deep = 0;
   make_tree($result,$result[count($result)-1],$deep);

}

function is_single_tag($test_tag)
{
   $single_tags=array("meta","img","link","input","!DOCTYPE","area","base","basefont","embed","hr","br");
   $single_tag_string="(";
   foreach($single_tags as $single_tag)
   {
       $single_tag_string=$single_tag_string.$single_tag."|" ;
   }
   $single_tag_string=$single_tag_string.")";
   $single_tag_string=str_replace("|)",")",$single_tag_string);
   $pattern="/\<".$single_tag_string.".*?\>/ i" ;

   if(preg_match($pattern,$test_tag))
   {
       return true ;
   }

   return false ;
}

function is_end_tag($test_tag)
{
   $pattern="/\<\/.*?\>/" ;
   if(preg_match($pattern,$test_tag))
   {
       return true ;
   }
   return false ;
}

function is_exist_start_tag($queue,$temptag)
{
   $end_string=str_replace("/","",$temptag);
   $end_string=preg_replace("/\s*/","",$end_string);
   $end_string=strtolower($end_string);

   $count=count($queue);

   echo $count."\n";

   var_dump($queue);

   for($i=$count-1;$i>=0;$i--)
   {

       echo $i."\n";
       $string = $queue[$i]["html_tag"];
       $string = preg_replace("/(<)(.*?)\s.*?(>)/",'$1$2$3',$string);
       if(strtolower($string) == strtolower($end_string))
       {
           return true;
       }
   }
   return false ;
}

function is_pear_tag($start_tag,$temptag)
{
   $start_string=preg_replace("/(<)(.*?)\s.*?(>)/",'$1$2$3',$start_tag);
   $end_string=str_replace("/","",$temptag);
   $end_string=preg_replace("/\s*/","",$end_string);
   if(strtolower($start_string) == strtolower($end_string))
   {
       return true ;
   }
   else
   {
       return false ;
   }
}

function make_string($deep)
{
   $temstr="";
   for($i=0;$i<$deep;$i++)
       $temstr=$temstr."-";
   return $temstr;
}

function judge_tag($last_tag,$temptag)
{
   return false;
// 这里做一个判断判断上一个标签是否能嵌套这个标签比如
// <td> 不嵌套<td>
// <td> 不嵌套<tr>
// <tr> 不嵌套<tr>
// <tr> 不嵌套<table>
// <table> 嵌套 <tr> 但不能嵌套其他标签
// <li> 不嵌套<li>
// <ul> 不嵌套<ul>
// <ol> 不嵌套<ol>
// <head> 不能嵌套 <body>
}

function get_content_by_url($site_url)
{
   $ch=curl_init($site_url);
   curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
   curl_setopt($ch, CURLOPT_TIMEOUT, 60);
   curl_setopt($ch, CURLOPT_CONNECTTIMEOUT, 10);
   curl_setopt($ch, CURLOPT_ENCODING, "gzip");
   $content=curl_exec($ch);
   $content = str_replace("\r\n","",$content);
   $content = str_replace("\n\r","",$content);
   $content = str_replace("\n","",$content);
   $content = str_replace("\r","",$content);
   $content = preg_replace("//is","",$content);
   $content = preg_replace("//is","",$content);
   $content = preg_replace("/<script(.*?)<\/script>/is","",$content);
   $content = preg_replace("/<style(.*?)<\/style>/is","",$content);

   $errores = fopen("baidu","w");
   fwrite($errores,$content) ;
   fclose($errores);

   return $content ;
}

function make_tree($result,$node,$deep)
{
   // 制作前缀使得显示的层次分明
   $str="" ;
   for($i=0;$i<$deep;$i++)
   {
       $str=$str."--";
   }
   echo $str;
   echo $node["html_tag"]."\n";

   $write_str = $str.$node["html_tag"]."\n";

   $tree_file = fopen("tree_file","a");
   fwrite($tree_file,$write_str);
   fclose($tree_file);

   $deep++;

   if(!empty($node["ChildId"]))
   {
       foreach($node["ChildId"] as $Child)
       {
           make_tree($result,$result[$Child],$deep) ;   ///递归
       }
   }
   else
   {
       return ;
   }
}
?>