php多线程爬虫类

代码：

]>
* @property
* 1、calltrigger    触发爬虫程序的回调函数
* 2、calltodo       处理业务逻辑的回调函数 如：把抓取到的内容处理后存到数据库
* 3、timeout        超时时间，默认5秒
* 4、depth          重定向深度，默认3
* 5、name           上传文件的名字，默认file
* 6、cookie         模拟登录时cookie存储在本地的文件，默认cookie_n.txt
* @method
* 1、ssl            是否设置https           true:是  false:否
* 2、auth           启用验证                user:用户名    pass:密码
* 3、login          模拟登录，获取cookie
* 4、cookie         使用cookie登录
* 5、header         设置请求头              data:请求头数组
* 6、proxy          设置服务器代理          url:代理服务器url   port:代理服务器端口
* 7、agent          设置浏览器代理          browse:代理浏览器 默认:Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko)
* 8、get            模拟get请求             data:传递的数据
* 9、post           模拟post请求            data:传递的数据
* 10、json          模拟json请求            data:传递的数据
* 11、upload        模拟表单上传            files:上传的文件   array|string
* 12、download      下载文件                dir:要下载的文件  格式：a/b
* 13、run           执行                    depth:深度
*/
class crawl{
public $calltrigger = 'trigger';  #  触发爬虫程序的回调函数
public $calltodo = 'todo';  #  处理业务逻辑的回调函数 
public $timeout = 5;  #  超时时间，默认5秒
public $depth = 3;  #  重定向深度，默认3
public $name = 'file';  #  上传文件的名字，默认file
public $cookie = 'cookie.txt';  #  模拟登录时cookie存储在本地的文件，默认cookie_n
private $schemes = array();
private $hosts = array();
private $paths = array();
private $querys = array();
private $options = array();
private $chs;
private $fps;
private $handle;
private $urls = array();
/*
 @desc：内部方法，获取页面中的超链接
 @param content 页面内容
 @return urls 获取到的超链接
 */
private function geturl($content){
    $preg = '/<[a|A].*?href=[\'\"]{0,1}([^>\'\"\ ]*).*?>/i';
    $bool = preg_match_all($preg,$content,$res);
    $urls = array();
    if($bool){
        $urls = $res[1];
    }
    $urls = array_unique($urls);
    return $urls;
}
/*
 @desc：内部方法，修复不完整的url
 @param url 原始url
 @param url 修复好的url
 */
private function reviseurl($url){
    $info = parse_url($url);
    $scheme = $info["scheme"]?:'http';
    $user = $info["user"];
    $pass = $info["pass"];
    $host = $info["host"];
    $port = $info["port"];
    $path = $info["path"];
    $url = $scheme . '://';
    if ($user && $pass) {
        $url .= $user . ":" . $pass . "@";
    }
    $url .= $host;
    if ($port) {
        $url .= ":" . $port;
    } 
    $url .= $path;
    return $url;
}
/*
 @desc：内部方法，调用回调函数进行业务处理
 @param content 传入到回调函数的参数
 */
private function todo($content){
    $calltodo = $this->calltodo;
    call_user_func($calltodo,$content);
}
/*
 @desc：触发爬虫程序的回调函数
 @param urls 待处理的url数组
 @param depth 处理深度
 */
private function trigger($urls,$depth){
    $calltrigger = $this->calltrigger;
    call_user_func($calltrigger,$urls,$depth);
}
/*
 @desc：内部方法 设置get请求参数
 @param data 请求数据
 */
private function setget($data){
    $schemes = $this->schemes;
    $hosts = $this->hosts;
    $paths = $this->paths;
    $querys = $this->querys;
    foreach($this->chs as $k=>$v){
        $sep = ($querys[$k] || !empty($data))?"?":"";
        $qurl = $schemes[$k].'://'.$hosts[$k].$paths[$k].$sep.$querys[$k].$data;
        $this->options[$k][CURLOPT_URL] = $qurl;
    }
    return $this;
}
/*
 @desc：内部方法 设置post请求参数
 @param data 请求数据
 */
private function setpost($data){
    $schemes = $this->schemes;
    $hosts = $this->hosts;
    $paths = $this->paths;
    $querys = $this->querys;
    foreach($this->chs as $k=>$v){
        $sep = $query?"?":"";
        $qurl = $schemes[$k].'://'.$hosts[$k].$paths[$k].$sep.$querys[$k];
        $this->options[$k][CURLOPT_URL] = $qurl;
        $this->options[$k][CURLOPT_POST] = 1;
        $this->options[$k][CURLOPT_POSTFIELDS] = $data;
    }
    return $this;
}
/*
 @desc：内部方法 设置最终请求参数
 */
private function setopt(){
    $options = $this->options;
    foreach($options as $k=>$v){
        curl_setopt_array(
                $this->chs[$k],
                $v
            );
    }
    return $this;
}
/*
 @desc：构造方法 设置初始请求参数
 @param urls 请求地址数组
 */
public function __construct($urls){
    $this->urls = $urls;
    $this->handle = curl_multi_init();
    foreach($urls as $k=>$v){
        $info = parse_url($v);
        $this->schemes[$k] = $info['scheme']?:'http';
        $this->hosts[$k] = $info['host'];
        $this->paths[$k] = $info['path'];
        $this->querys[$k] = $info['query'];
        $this->chs[$k] = curl_init();
        $this->options[$k][CURLOPT_CONNECTTIMEOUT] = $this->timeout;
        $this->options[$k][CURLOPT_RETURNTRANSFER] = 1;
        $this->options[$k][CURLOPT_FOLLOWLOCATION] = 1;
        $this->options[$k][CURLINFO_HEADER_OUT] = true;
        $this->options[$k][CURLOPT_ENCODING] = 'gzip';
        $this->options[$k][CURLOPT_MAXREDIRS] = $this->depth;
        curl_multi_add_handle ($this->handle,$this->chs[$k]);
    }
}
/*
 @desc：是否设置https请求
 @param bool true:https请求 false:http请求
 */
public function ssl($bool = false){
    if($bool){
        foreach($this->chs as $k=>$v){
            $this->scheme[$k] = 'https';
            $this->options[$k][CURLOPT_SSL_VERIFYHOST] = 1;
            $this->options[$k][CURLOPT_SSL_VERIFYPEER] = false;
        }
    }
    return $this;
}
/*
 @desc：设置验证用户名、密码
 @param user 用户名
 @param pass 密码
 */
public function auth($user,$pass){
    foreach($this->chs as $k=>$v){
        $this->options[$k][CURLOPT_USERPWD] = $user.':'.$pass;
    }
    return $this;
}
/*
 @desc：模拟登录
 */
public function login(){
    $cookie = $this->cookie;
    $arr = explode('.',$cookie);
    $name = $arr[0];
    $ext = $arr[1];
    foreach($this->chs as $k=>$v){
        $this->options[$k][CURLOPT_COOKIEJAR] = $name.'_'.$k.'.'.$ext;
        $this->options[$k][CURLOPT_RETURNTRANSFER] = 0;
    }
    return $this;
}
/*
 @desc：带cookie登录
 */
public function cookie(){
    $cookie = $this->cookie;
    $arr = explode('.',$cookie);
    $name = $arr[0];
    $ext = $arr[1];
    foreach($this->chs as $k=>$v){
        $this->options[$k][CURLOPT_COOKIEFILE] = $name.'_'.$k.'.'.$ext;
    }
    return $this;
}
/*
 @desc：设置请求头信息
 @param data 请求头
 */
public function header($data){
    foreach($this->chs as $k=>$v){
        $this->options[$k][CURLOPT_HTTPHEADER] = $this->options[$k][CURLOPT_HTTPHEADER]?:array();
        $this->options[$k][CURLOPT_HTTPHEADER] = array_merge($this->options[$k][CURLOPT_HTTPHEADER],$data);
    }
    return $this;
}
/*
 @desc：设置代理服务器
 @param url 代理服务器url
 @param port 代理服务器端口
 */
public function proxy($url,$port){
    $info = parse_url($url);
    $scheme = $info['scheme']?:'http';
    $host = $info['host'];
    $path = $info['path'];
    $purl = $scheme.'://'.$host.$path.':'.$port;
    foreach($this->chs as $k=>$v){
        $this->options[$k][CURLOPT_PROXY] = $purl;
    }
    return $this;
}
/*
 @desc：设置代理浏览器
 @param browse 代理浏览器
 */
public function agent($browse = 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko)'){
    foreach($this->chs as $k=>$v){
        $this->options[$k][CURLOPT_USERAGENT] = $browse;
    }
    return $this;
}
/*
 @desc：模拟get请求
 @param data 请求数据
 */
public function get($data = array()){
    $data = http_build_query($data);
    $this->setget($data);
    return $this;
}
/*
 @desc：模拟post请求
 @param data 请求数据
 */
public function post($data = array()){
    $this->setpost($data);
    return $this;
}
/*
 @desc：模拟json请求
 @param data 请求数据
 */
public function json($data = array()){
    $data = json_encode($data);
    $header = array(
            'Content-Type: application/json',
            'Content-Length:' . strlen($data)
        );
    $this->header($header);
    $this->setpost($data);
    return $this;
}
/*
 @desc：模拟表单上传
 @param files 文件路径
 */
public function upload($files){
    $data = array();
    $name = $this->name;
    if(is_array($files)){
        foreach($files as $k=>$v){
            $data["{$name}[{$k}]"]=new \CURLFile($v);
        }
    }else{
        $data["{$name}"]=new \CURLFile($files);
    }
    $this->setpost($data);
    return $this;
}
/*
 @desc：下载文件
 @param dir 存储文件目录
 */
public function download($dir = ''){
    $paths = $this->paths;
    if($dir && !is_dir($dir)){
        mkdir($dir,0755,true);
    }
    foreach($this->paths as $k=>$v){
        $name = strrchr($v, '/');
        $dsep = $dir?'/':'';
        $this->fps[$k]=fopen('.'.$dsep.$dir.$name, 'w');
        $this->options[$k][CURLOPT_FILE] = $this->fps[$k];
    }
    $this->setget('');
    return $this;
}
/*
 @desc：执行方法
 @param depth 深度 默认2
 */
public function run($depth = 2){
    $this->setopt();
    $chs = $this->chs;
    $handle = $this->handle;
    $urls = $this->urls;
    if($depth > 0){
        $depth--;
        $active = null;
        $mrc = curl_multi_exec($handle, $active);
        while ($mrc == CURLM_CALL_MULTI_PERFORM) {
            $mrc = curl_multi_exec($handle, $active);
        }
        while ($active && $mrc == CURLM_OK) {
            if (curl_multi_select($handle) != -1) {  
                usleep(100);
            }
            $mrc = curl_multi_exec($handle, $active);
            while ($mrc == CURLM_CALL_MULTI_PERFORM) {
                $mrc = curl_multi_exec($handle, $active);
            }
        }
        foreach ($chs as $k => $v) {
            if (curl_error($chs[$k]) == "") {
                $content = curl_multi_getcontent($chs[$k]);
                $this->todo($content);
                $aurls = $this->geturl($content);
                $urls[$k] = $this->reviseurl($urls[$k]);
                if (is_array($aurls) && !empty($aurls)) {
                    foreach ($aurls as $k1=>$u) {
                        if (preg_match('/^http/', $u)) {
                            $returl[$k1] = $u;
                        } else {
                            $real = $urls[$k] . '/' . $u;
                            $returl[$k1] = $real;
                        }
                    }
                    $this->trigger($returl,$depth);
                }
            }
            curl_multi_remove_handle($handle, $chs[$k]);  
            curl_close($chs[$k]);
        }
        curl_multi_close($handle);
    }
}
}

测试：

function todo($content){
echo 'ok'.PHP_EOL;
}
$urls=array(
'www.baidu.com',  
'www.taobao.com'
);
function trigger($urls = array(),$depth = 2){
$crawl = new crawl($urls);
$crawl->get()->run($depth);
}
trigger($urls);

输出：

ok
ok
ok
ok
ok
ok
ok
ok
ok
ok
ok
ok
ok
ok

当前名称：php多线程爬虫类
转载来源：http://hbruida.cn/article/gojios.html

php多线程爬虫类

其他资讯