文章采集 正则版 2009-12-25

//采集html
function getwebcontent($url){
    $ch = curl_init();
    $timeout = 10;
    curl_setopt($ch, CURLOPT_URL, $url);
    curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
    curl_setopt($ch, CURLOPT_CONNECTTIMEOUT, $timeout);
    curl_setopt ($ch, CURLOPT_FOLLOWLOCATION, 1);
    $contents = trim(curl_exec($ch));
    curl_close($ch);
    return  $contents;
}
 
 
//获得标题和url
$string = 
 getwebcontent('http://www.babytree.com/learn/zhunbeihuaiyun/jijibeiyun/2');
//正则匹配<li>获取标题和地址
preg_match_all ("/<li><a href=\"\/learn\/article\/(.*)\">(.*)<\/a>/",$string, $out, PREG_SET_ORDER);
foreach($out as $key => $value){
    $article['title'][]   = $out[$key][2];
    $article['link'][]    = "http://www.babytree.com/learn/article/".$out[$key][1];
}
//根据url获取文章内容
foreach($article['link'] as $key=>$value){
    $content_html = getwebcontent($article['link'][$key]);
    preg_match("/<div id=pagenum_0(.*)>[\s|\S]*?<\/div>/",$content_html,$matches);
    $article[content][$key] = $matches[0];
 
}
//不转码还真不能保存成文件
foreach($article[title] as $key=>$value){
$article[title][$key] = iconv('utf-8', 'gbk', $value);//转码
}
//存入文件
$num = count($article['title']);
for($i=0; $i<$num; $i++){
     file_put_contents("{$article[title][$i]}.txt", $article['content'][$i]);
}
?>

特别推荐 这个小哥的博客可以小学一下正则表达式，语言还很风趣，呵呵 686条回复啊！ 强力推荐！

http://www.cnblogs.com/deerchao/archive/2006/08/24/zhengzhe30fengzhongjiaocheng.html