代码如下:
<?php
$oldSetting = libxml_use_internal_errors( true );
libxml_clear_errors();
/**
*
* -+-----------------------------------
* |PHP5 Framework - 2011
* |Web Site: www.iblue.cc
* |E-mail: mejinke@gmail.com
* |Date: 2012-10-12
* -+-----------------------------------
*
* @desc HTML解析器
* @author jingke
*/
class XF_HtmlDom
{
private $_xpath = null;
private $_nodePath = '';
public function __construct($xpath = null, $nodePath = '')
{
$this->_xpath = $xpath;
$this->_nodePath = $nodePath;
}
public function loadHtml($url)
{
ini_set('user_agent', 'Mozilla/5.0 (Linux; U; Android 2.1; en-us; Nexus One Build/ERD62) AppleWebKit/530.17 (KHTML, like Gecko) Version/4.0 Mobile Safari/530.17 –Nexus');
$content = '';
if(strpos(strtolower($url), 'http')===false)
{
$content = file_get_contents($url);
}
else
{
$ch = curl_init();
$user_agent = "Baiduspider+(+http://www.baidu.com/search/spider.htm)";
$user_agent1='Mozilla/5.0 (Windows NT 5.1; rv:6.0) Gecko/20100101 Firefox/6.0';
curl_setopt($ch, CURLOPT_URL, $url);
curl_setopt($ch, CURLOPT_HEADER, false);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
curl_setopt($ch, CURLOPT_REFERER, $url);
curl_setopt($ch, CURLOPT_USERAGENT, $user_agent1);
curl_setopt($ch, CURLOPT_FOLLOWLOCATION,1);
