使用PHP切割HTML页面DIV

时间:2014-07-22 14:50来源: 作者: 点击: 次

分享到：

首次发表自博客园-Zjmainstay： a href=http://www.cnblogs.com/Zjmainstay/archive/2012/08/06/php_getDivContain.html rel=nofollowphp获取页面并切割页面div内容/a br / br /

首次发表自博客园-Zjmainstay： php获取页面并切割页面div内容

<?php
    header("Content-type: text/html; charset=utf-8"); 
    function getWebDiv($div_id,$url=false,$data=false){
        if($url !== false){
            $data = file_get_contents( $url );
        }
        $charset_pos = stripos($data,'charset');
        if($charset_pos) {
            if(stripos($data,'utf-8',$charset_pos)) {
                $data = iconv('utf-8','utf-8',$data);
            }else if(stripos($data,'gb2312',$charset_pos)) {
                $data = iconv('gb2312','utf-8',$data);
            }else if(stripos($data,'gbk',$charset_pos)) {
                $data = iconv('gbk','utf-8',$data);
            }
        }
        //获取所有div前缀
        preg_match_all('/<div/i',$data,$pre_matches,PREG_OFFSET_CAPTURE);
        //获取所有div后缀
        preg_match_all('/<\/div/i',$data,$suf_matches,PREG_OFFSET_CAPTURE); 
        $hit = strpos($data,$div_id);
        if($hit == -1) return false;    //未命中
        $divs = array();    //合并所有div
        foreach($pre_matches[0] as $index=>$pre_div){
            $divs[(int)$pre_div[1]] = 'p';
            $divs[(int)$suf_matches[0][$index][1]] = 's';    
        }
        
        //对div进行排序
        $sort = array_keys($divs);
        asort($sort);
        
        $count = count($pre_matches[0]);
        foreach($pre_matches[0] as $index=>$pre_div){
            //<div $hit <div+1    时div被命中
            if(($pre_matches[0][$index][1] < $hit) 
                && ($hit < $pre_matches[0][$index+1][1])){
                $deeper = 0;
                //弹出被命中div前的div
                while(array_shift($sort) != $pre_matches[0][$index][1] 
                    && ($count--)) continue;
                //对剩余div进行匹配，若下一个为前缀，则向下一层，$deeper加1，
                //否则后退一层，$deeper减1，$deeper为0则命中匹配，计算div长度
                foreach($sort as $key){
                    if($divs[$key] == 'p') $deeper++;
                    else if($deeper == 0) {
                        $length = $key-$pre_matches[0][$index][1];
                        break;
                    }else {
                        $deeper--;
                    }
                }
                $hitDivString = substr($data,$pre_matches[0][$index][1],$length).'</div>';
                break;
            }
        }
        return $hitDivString;
    }
    //考虑到id单双引号符号问题，id="u"由用户自己填写。
    //声明：此段php只针对带 id div内容的读取。
    echo getWebDiv('id="taglist"','http://www.cnblogs.com/Zjmainstay/tag/');

//End_php

2. [代码]完善：匹配任意闭合带id标签跳至 [1] [2] [全屏预览]

<?php
    header("Content-type: text/html; charset=utf-8"); 
    function getWebTag($tag_id,$url=false,$tag='div',$data=false){
        if($url !== false){
            $data = file_get_contents( $url );
        }
        $charset_pos = stripos($data,'charset');
        if($charset_pos) {
            if(stripos($data,'utf-8',$charset_pos)) {
                $data = iconv('utf-8','utf-8',$data);
            }else if(stripos($data,'gb2312',$charset_pos)) {
                $data = iconv('gb2312','utf-8',$data);
            }else if(stripos($data,'gbk',$charset_pos)) {
                $data = iconv('gbk','utf-8',$data);
            }
        }
        
        preg_match_all('/<'.$tag.'/i',$data,$pre_matches,PREG_OFFSET_CAPTURE);    //获取所有div前缀
        preg_match_all('/<\/'.$tag.'/i',$data,$suf_matches,PREG_OFFSET_CAPTURE); //获取所有div后缀
        $hit = strpos($data,$tag_id);
        if($hit == -1) return false;    //未命中
        $divs = array();    //合并所有div
        foreach($pre_matches[0] as $index=>$pre_div){
            $divs[(int)$pre_div[1]] = 'p';
            $divs[(int)$suf_matches[0][$index][1]] = 's';    
        }
        
        //对div进行排序
        $sort = array_keys($divs);
        asort($sort);
        
        $count = count($pre_matches[0]);
        foreach($pre_matches[0] as $index=>$pre_div){
            //<div $hit <div+1    时div被命中
            if(($pre_matches[0][$index][1] < $hit) && ($hit < $pre_matches[0][$index+1][1])){
                $deeper = 0;
                //弹出被命中div前的div
                while(array_shift($sort) != $pre_matches[0][$index][1] && ($count--)) continue;
                //对剩余div进行匹配，若下一个为前缀，则向下一层，$deeper加1，
                //否则后退一层，$deeper减1，$deeper为0则命中匹配，计算div长度
                foreach($sort as $key){
                    if($divs[$key] == 'p') $deeper++;
                    else if($deeper == 0) {
                        $length = $key-$pre_matches[0][$index][1];
                        break;
                    }else {
                        $deeper--;
                    }
                }
                $hitDivString = substr($data,$pre_matches[0][$index][1],$length).'</'.$tag.'>';
                break;
            }
        }
        return $hitDivString;
    }
    
    echo getWebTag('id="nav"','http://mail.163.com/html/mail_intro/','ul');
    echo getWebTag('id="homeBanners"','http://mail.163.com/html/mail_intro/');
    echo getWebTag('id="performance"','http://mail.163.com/html/mail_intro/','section');

//End_php