php 利用curl，正则表达式做的一个php蜘蛛抓取器

凤网fcms内容管理系统get.php 抓取框架，对网页内容的分析处理并进行相关替换std.php 通用正则news_67_com.php 对http://news.67.com 的抓取分析器先抓列表，再抓内容页。还欠缺监控，统计，错误处理功能。个人觉得还是比较好玩。

news_67_com.php

<?php
include_once dirname(__FILE__) . '/std.php';

$site = array(
    'aname'     => '中国娱乐网',
    'domain'    => 'news.67.com',
    'dirname'   => '目录名称，用于匹配基于目录不同的正文',
    'gettype'   => 'default',
    //获取主文件
    'creg'      => '/(?si)<!--文章 begin-->(.*?)\<\!--文章 end-->/',
    'code'      => 'utf-8',
    'sub'       => '获取子目录正则',
    'content'   => 'tag1',
    'img_upload'=> array('tag1' => ''),
    //下一页
    'reg_next'  => '/(?is)<a target=\'_self\' href=\'([^\']*?)\'>下一页\>\><\/a>/',
    'key0'      => '/(?is)<meta name="keywords" content="([^"]*?)".*?\/>/',
    'key0_ap'   => array(array(',', '|'), ' '),
    'tag0'      => '/(?is)<h1>([^<^>]*?)<\/h1>/',
    'tag0_arp'  => array(
        array(
            '/(?is)\(组图\)/',
            '/(?is)\(图\)/',
            '/(?is)\(图\.\./',
            '/(?is)\(组图\.\./',
            '/(?is)\./',
            '/(?is)(《|》)/',
        ),
        array(
            '', '', '', '', '', '',
        )
    ),
    'tag1'      => '/(?is)<div class="article" id="divContent">(.*?)<img class="[^"]*?" style="[^"]*?" src="[^"]*?" alt="[^"]*?" border="\d*" \/>/',
    'tag1_brp'  => array(
        array(
            '/(?is)（.*?）/',
            '/(?is)\(.*?\)/',
            '/(?is)\s*<p align="center">.*?<img.*?src="([^"]*?)".*?>(.*?)<\/p>\s*/',
            '/(?is)\s*<p>\s*/',
            '/(?is)\s*<p align="center">\s*/',
            '/(?is)　/',
            '/(?is)<br \/>/',
            '/(?is)\s*<p align="left">\s*/',
            '/(?is)\s*<p class="f_center" align="center">\s*/',
            '/(?is)\s*<center>\s*/',
            '/(?is)\s*<\/center>\s*/',
            '/(?is)\s*<p class="f_center">\s*/',
        ),
        array(
            '', '', '<p style="text-align: center;"><img src="$1" /></p>', '<p style="text-indent: 24px;">', '<p style="text-align: center;">', '', '', '<p style="text-indent: 24px;">', '<p style="text-align: center;">', '<p style="text-align: center;">', '</p>', '<p style="text-indent: 24px;">'
        ),
    ),
    'tag1_arp'  => array(
        array(
            '/(?is)<p style="text-align: center;"> <\/p>/',
            '/(?is)<strong><\/strong>/'
        ),
        array(
            '', ''
        ),
    ),
    'strip'     => array('tag1' => ''),
    'tag2'      => '/(?is)<div class="daodu">导读：\s*(.*?)\s*<\/div>/',
    'tag2_arp'  => array(
        array(
            '/(?is)　/'
        ),
        array(
            ''
        ),
    ),
    'tag3'      => '/(?is)(中国娱乐网)/',
    'tag4'      => '/(?is)<div class="artInfo"><span>日期：(\d+-\d+-\d+ \d+:\d+:\d+).*?<\/div>/',
);

$map = array(
    'tag'       => 'key0',
    'title'     => 'tag0',
    'content'   => 'tag1',
    'summary'   => 'tag2',
    'source'    => 'tag3',
    'pub_date'  => 'tag4',
);

$site_list = array(
    'aname'     => '中国娱乐网',
    'domain'    => 'www.67.com',
    'gettype'   => 'default',
    'creg'      => '/(?si)<div class="gallery_list">(.*?)<div class="nt_cl">/',
    'code'      => 'gbk',
    'reg_next'  => '/(?si)<li class="next"><a href="([^"]+?)" target="_self">下一页<\/a><\/li>/',
    //链接
    'tag0'      => '/(?is)<div style="height: 30px;">.*?<a target="_blank"\s*href=\'(\w+:\/\/news\.67\.com\/\w+\/\d+\/\d+\/\d+\/\d+\.\w+)\s*\' style="font-size: 14px;">[^<^>]*?<\/a>.*?<\/div>/',
    //标题
    'tag1'      => '/(?is)<div style="height: 30px;">.*?<a target="_blank"\s*href=\'\w+:\/\/news\.67\.com\/\w+\/\d+\/\d+\/\d+\/\d+\.\w+\s*\' style="font-size: 14px;">([^<^>]*?)<\/a>.*?<\/div>/',
    'tag1_arp'  => array(
        array(
            '/(?is)\(组图\)/',
            '/(?is)\(图\)/',
            '/(?is)\(图\.\./',
            '/(?is)\(组图\.\./',
            '/(?is)\./',
            '/(?is)(《|》)/',
        ),
        array(
            '', '', '', '', '', '',
        )
    ),
);

$list_map = array(
    'url'       => 'tag0',
    'title'     => 'tag1',
);

$site_list_sub = array();