PHP利用Curl实现多线程抓取网页和下载文件
PHP 利用 Curl 可以完成各种传送文件操作,比如模拟浏览器发送GET,POST请求等等,然而因为php语言本身不支持多线程,所以开发爬虫程序效率并不高,一般采集数据可以利用 PHPquery类来采集数据库,在此之外也可以用 Curl ,借助Curl 这个功能实现并发多线程的访问多个url地址以实现并发多线程抓取网页或者下载文件.
至于具体实现过程,请参考下面几个例子:
$urls = array( <br />'http://www.scutephp.com/', <br />'http://www.google.com/', <br />'http://www.example.com/' <br />); // 设置要抓取的页面URL <br />$save_to='/test.txt'; // 把抓取的代码写入该文件 <br />$st = fopen($save_to,"a"); <br />$mh = curl_multi_init(); <br />foreach ($urls as $i => $url) { <br />$conn[$i] = curl_init($url); <br />curl_setopt($conn[$i], CURLOPT_USERAGENT, "Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 6.0)"); <br />curl_setopt($conn[$i], CURLOPT_HEADER ,0); <br />curl_setopt($conn[$i], CURLOPT_CONNECTTIMEOUT,60); <br />curl_setopt($conn[$i], CURLOPT_FILE,$st); // 将爬取的代码写入文件 <br />curl_multi_add_handle ($mh,$conn[$i]); <br />} // 初始化 <br />do { <br />curl_multi_exec($mh,$active); <br />} while ($active); // 执行 <br />foreach ($urls as $i => $url) { <br />curl_multi_remove_handle($mh,$conn[$i]); <br />curl_close($conn[$i]); <br />} // 结束清理 <br />curl_multi_close($mh); <br />fclose($st);
下面这段代码和上面差不多意思,只不过这个地方是将获得的代码先放入变量,然后再将获取到的内容写入指定的文件
$urls = array( <br />'http://www.scutephp.com/', <br />'http://www.google.com/', <br />'http://www.example.com/' <br />); <br />$save_to='/test.txt'; // 把抓取的代码写入该文件 <br />$st = fopen($save_to,"a"); <br />$mh = curl_multi_init(); <br />foreach ($urls as $i => $url) { <br />$conn[$i] = curl_init($url); <br />curl_setopt($conn[$i], CURLOPT_USERAGENT, "Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 6.0)"); <br />curl_setopt($conn[$i], CURLOPT_HEADER ,0); <br />curl_setopt($conn[$i], CURLOPT_CONNECTTIMEOUT,60); <br />curl_setopt($conn[$i],CURLOPT_RETURNTRANSFER,true); // 不将爬取代码写到浏览器,而是转化为字符串 <br />curl_multi_add_handle ($mh,$conn[$i]); <br />} <br />do { <br />curl_multi_exec($mh,$active); <br />} while ($active); <br />foreach ($urls as $i => $url) { <br />$data = curl_multi_getcontent($conn[$i]); // 获得爬取的代码字符串 <br />fwrite($st,$data); // 将字符串写入文件<br />} // 获得数据变量,并写入文件 <br />foreach ($urls as $i => $url) { <br />curl_multi_remove_handle($mh,$conn[$i]); <br />curl_close($conn[$i]); <br />}<br />curl_multi_close($mh); <br />fclose($st);
$urls=array(<br />'http://www.scutephp.com/5w.zip',<br />'http://www.scutephp.com/5w.zip',<br />'http://www.scutephp.com/5w.zip'<br />);<br />$save_to='./home/';<br />$mh=curl_multi_init();<br />foreach($urls as $i=>$url){<br />$g=$save_to.basename($url);<br />if(!is_file($g)){<br />$conn[$i]=curl_init($url);<br />$fp[$i]=fopen($g,"w");<br />curl_setopt($conn[$i],CURLOPT_USERAGENT,"Mozilla/4.0(compatible; MSIE 7.0; Windows NT 6.0)");<br />curl_setopt($conn[$i],CURLOPT_FILE,$fp[$i]);<br />curl_setopt($conn[$i],CURLOPT_HEADER ,0);<br />curl_setopt($conn[$i],CURLOPT_CONNECTTIMEOUT,60);<br />curl_multi_add_handle($mh,$conn[$i]);<br />}<br />}<br />do{<br />$n=curl_multi_exec($mh,$active);<br />}while($active);<br />foreach($urls as $i=>$url){<br />curl_multi_remove_handle($mh,$conn[$i]);<br />curl_close($conn[$i]);<br />fclose($fp[$i]);<br />}<br />curl_multi_close($mh);$urls=array(<br />'http://www.scutephp.com/5w.zip',<br />'http://www.scutephp.com/5w.zip',<br />'http://www.scutephp.com/5w.zip'<br />);<br />$save_to='./home/';<br />$mh=curl_multi_init();<br />foreach($urls as $i=>$url){<br />$g=$save_to.basename($url);<br />if(!is_file($g)){<br />$conn[$i]=curl_init($url);<br />$fp[$i]=fopen($g,"w");<br />curl_setopt($conn[$i],CURLOPT_USERAGENT,"Mozilla/4.0(compatible; MSIE 7.0; Windows NT 6.0)");<br />curl_setopt($conn[$i],CURLOPT_FILE,$fp[$i]);<br />curl_setopt($conn[$i],CURLOPT_HEADER ,0);<br />curl_setopt($conn[$i],CURLOPT_CONNECTTIMEOUT,60);<br />curl_multi_add_handle($mh,$conn[$i]);<br />}<br />}<br />do{<br />$n=curl_multi_exec($mh,$active);<br />}while($active);<br />foreach($urls as $i=>$url){<br />curl_multi_remove_handle($mh,$conn[$i]);<br />curl_close($conn[$i]);<br />fclose($fp[$i]);<br />}<br />curl_multi_close($mh);