本帖最后由 oasisxp 于 2014-08-25 13:45:08 编辑 想用PHP的CURL采集虾米网的音乐信息。
$j=0;<br /> //起始ID<br /> $id = 200000;<br /> //采集1000条<br /> //保存采集的数据<br /> $data = array();<br /> while($j<1000){<br /> $url = 'http://www.xiami.com/song/'.($id++);<br /> $ch = curl_init();<br /> $status = curl_getinfo($ch);<br /> ///$status['redirect_url'] ;// 跳转到的新地址<br /> $header[]='Accept:text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8';<br /> $header[]='Accept-Encoding:gzip,deflate,sdch';<br /> $header[]='Accept-Language:zh-CN,zh;q=0.8';<br /> $header[]='Cache-Control:max-age=0';<br /> $header[]='Connection:keep-alive';<br /> $header[]='Cookie:_unsign_token=a35437bd35c221c09a0e6f564e17c225; __gads=ID=7fcc242f6fd63d77:T=1408774454:S=ALNI_Mae8MH6vL5z6q4NlGYzyqgD4jHeEg; bdshare_firstime=1408774454639; _xiamitoken=3541aab48832ba3ceb089de7f39b9b0f; pnm_cku822=211n%2BqZ9mgNqgJnCG0Zu8%2BzyLTPuc%2B7wbrff98%3D%7CnOiH84T3jPCG%2FIr%2BiPOG8lI%3D%7CneiHGXz6UeRW5k4rRCFXIkcoTdd7ym3fZdO2FrY%3D%7Cmu6b9JHlkuGa5pDqnOie5ZDkmeqb4ZTule6V7ZjjlOib7JrmkvdX%7Cm%2B%2BT%2FGIUew96DXsUYBd4HawbrTOXOVI4iyOLIYUqT%2B9P%7CmO6BH2wDcB9rHGsYdwRrH2gfbAN%2FDH8QZBNkF3gDeQqqCg%3D%3D%7Cme6d7oHyneiH84Twn%2BmR64TzUw%3D%3D; CNZZDATA921634=cnzz_eid%3D1437506062-1408774274-%26ntime%3D1408937320; CNZZDATA2629111=cnzz_eid%3D2021816723-1408774274-%26ntime%3D1408937320; isg=075E6FBDF77039CEB63A1BA239420244';<br /> $header[]='Host:www.xiami.com';<br /> $header[]='User-Agent:Mozilla/5.0 (Windows NT 5.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1653.0 Safari/537.36';<br /><br /> curl_setopt($ch, CURLOPT_URL, $url); //要访问的地址<br /> curl_setopt($ch, CURLOPT_HTTPHEADER, $header); //设置http头<br /> curl_setopt($ch, CURLOPT_HEADER, 0); //显示返回的Header区域内容<br /> curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1); //获取的信息以文件流的形式返回<br /> curl_setopt($ch, CURLOPT_TIMEOUT, 20); //设置超时限制防止死循环<br /> $content = curl_exec($ch); //执行操作<br /> $curl_errno = curl_errno($ch);<br /> $curl_error = curl_error($ch);<br /> curl_close($ch); //关闭CURL会话<br /> preg_match('/name="description"\s+content="《(.+)》演唱者(.+),所属专辑《(.+)》/', $content,$matches);<br /> //如果歌曲名字为空,跳过<br /> if(empty($matches[1]) || trim($matches[1]) == ''){<br /> continue;<br /> }<br /> <br /> //匹配出的数据<br /> $data[$id]['song'] = empty($matches[1])?' ':$matches[1];<br /> $data[$id]['songer'] = empty($matches[2])?' ':$matches[2];<br /> $data[$id]['album'] = empty($matches[3])?' ':$matches[3];<br /> <br /> preg_match('/album\/(\d+)/', $content,$matches);<br /><br /> $data[$id]['albumId'] = empty($matches[1])?0:$matches[1];<br /><br /> preg_match('/\/artist\/(\d+)/', $content,$matches);<br /> $data[$id]['songerId'] = empty($matches[1])?0:$matches[1];<br /><br /> //歌词<div class="lrc_main"><br /> preg_match('/<div class="lrc_main">(.*)<\/div>/Us', $content,$matches);<br /> $data[$id]['lrc'] = empty($matches[1])?' ':addslashes($matches[1]);<br /> //分享 分享<em>(3269)</em><br /> preg_match('/分享<em>\((\d+)\)<\/em>/Us', $content,$matches);<br /> $data[$id]['share'] = empty($matches[1]) ? 0:$matches[1];<br /> //评论次数 <p class="wall_list_count"><span>920<br /> preg_match('/<p class="wall_list_count"><span>(\d+)<\/span>/Us', $content,$matches);<br /> $data[$id]['comment_count'] = empty($matches[1])?0:$matches[1];<br /><br /><br /> //入库操作<br /> //print_r($data);<br /> //_____________________________<br /> $j++;<br /> usleep(3000);<br /> }
Salin selepas log masuk
本帖最后由 PhpNewnew 于 2014-08-27 22:09:20 编辑 亲用 Ruby 或者 Go 吧
开玩笑,就算你要跑好歹你也弄成命令行的模式跑呀....
------解决方案--------------------应该是xiami.com服务器有限制,禁止采集吧
1,每个url请求只采10-20打,然后做个跳转在继续采集,这样也可以防止页面超时,如果你在虚机上运行,长时间点用cpu,进程可能会被kill.
2,每次url请求header中的user-agent,cookies,最好都能改一下。
3,如果还不行,用火车头试试吧!
4,如果火车也不行,那就放弃这个站吧!
------解决方案--------------------把foreach拆分成循环执行同一页面。
第一次浏览器或者cronrab定时执行 http://localhost/caiji.php?num=1 每次完成后,$_GET['num']+1;curl 重复l执行同一脚本,当$_GET['num']==1000后,退出,不再执行curl。
if($_GET['num']){<br />$url = 'http://www.xiami.com/song/'.$_GET['num'];<br />//你的代码<br />$_GET['num'])++;<br />}<br />if($_GET['num']<1001){<br /> $ch = curl_init();<br /> curl_setopt($ch, CURLOPT_URL,"http://localhost/caiji.php?num=".$_GET['num']));<br /> curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);<br /> curl_setopt($ch, CURLOPT_CONNECTTIMEOUT ,2); <br /> curl_setopt($ch, CURLOPT_TIMEOUT ,2);<br /> curl_exec($ch);<br /> curl_close($ch);<br />}else{<br /> exit;<br />}<br />
Salin selepas log masuk