private function analysis_html($page, $loction_url)
{
$spider_url = array();
//spider_url表的数组
$spider_url['content'] = NULL;
//抽取正文内容
$iTextExtractor = new textExtract($page, 6);
//$spider_url['content'] = $iTextExtractor->getPlainText();
$s_text = mb_substr($iTextExtractor->getPlainText(), 0, 1000);
$spider_url['content'] = $s_text;
/*
$Readability = new Readability($page); // default charset is utf-8
$ReadabilityData = $Readability->getContent();
$spider_url['content'] = strip_tags($ReadabilityData['content'],'<img>');
*/
//先把header 用正则提取出来
preg_match("@<head[^>]*>(.*?)<\\/head>@si", $page, $regs);
$headdata = $regs[1];
//头部
$spider_url['description'] = NULL;
$spider_url['keywords'] = NULL;
$spider_url['author'] = NULL;
$spider_url['title'] = NULL;
$res = array();
if (isset($headdata)) {
#获取Description
preg_match("/<meta +name *=[\"']?description[\"']? *content=[\"']?([^<>'\"]+)[\"']?/i", $headdata, $res);
if (isset($res)) {
$spider_url['description'] = $res[1];
}
//获取keywords
preg_match("/<meta +name *=[\"']?keywords[\"']? *content=[\"']?([^<>'\"]+)[\"']?/i", $headdata, $res);
if (isset($res)) {
$spider_url['keywords'] = $res[1];
}
//获取author
preg_match("/<meta +name *=[\"']?author[\"']? *content=[\"']?([^<>'\"]+)[\"']?/i", $headdata, $res);
if (isset($res)) {
@($spider_url['author'] = $res[1]);
}
//title
preg_match("/<title>(.*)<\\/title>/Ui", $headdata, $res);
if (isset($res)) {
$spider_url['title'] = trim($res[1]);
}
}
//echo md5($page);
$spider_url['url'] = $loction_url;
$spider_url['content_md5'] = md5($page);
$spider_url['addtime'] = date("Y-m-d H:m:s");
$spider_url['site_id'] = $this->site_id;
//把数组写入或更新到spider_url表
$result['add_spider_url'] = $this->CI->spider_model->add_spider_url($spider_url);
return $result;
}
clear: both;
font-size: 12px;
}
</style>
</head>
<body>
<center>
<div id="allcontent">
<div id="title"><p>网页正文提取演示</p></div>
<?php
if (isset($_POST['url'])) {
set_time_limit(60 * 10);
require_once 'class.textExtract.php';
$iTextExtractor = new textExtract($_POST['url']);
$text = $iTextExtractor->getPlainText();
if ($iTextExtractor->isGB) {
$text = iconv('GBK', 'UTF-8//IGNORE', $text);
}
echo '<form method="post" action="demo.php">
<span class="des">网址：</span><input type="text" name="url" size="60" />
<input type="submit" name="submit" value="提取" />
</form>';
echo '<p id="gap"><?p>';
echo '<p id="lf" class="des">正文：</p>';
echo '<p id="text">' . $text . '</p>';
} else {
echo '<form method="post" action="demo.php">
<span class="des">网址：</span><input type="text" name="url" size="60" />
<input type="submit" name="submit" value="提取" />
</form>';