crm/extend/douyin/VideoCrawler.php

<?php

namespace douyin;

class VideoCrawler
{


    public function crawByUrl($url)
    {
        header('Access-Control-Allow-Origin:*');
        header('Content-type:application/json; charset=utf-8');
        $forwardId = explode('/', parse_url($this->getrealurl($url))['path'])[2];
        do {
            $ids = $forwardId;
            $json = json_decode($this->get_curl('https://www.iesdouyin.com/web/api/v2/aweme/iteminfo/?item_ids=' . $ids));
            if(count($json->item_list) == 0) return false;
            $json = $json->item_list[0];
            $forwardId = $json->forward_id;
        } while ($forwardId != 0);

        // 描述获取
        $desc = $json->share_info->share_weibo_desc;
        $playurl = $json->video->play_addr->url_list[0];
        $play =  $this->get_curl(str_ireplace('playwm', 'play', $json->video->play_addr->url_list[0]), ['ua' => 'Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/75.0.3770.100 Mobile Safari/537.36']);
        if(preg_match_all("/<a[^<>]+href *= *[\"']?(http\:\/\/[^ '\"]+)/i", $play, $body_links, PREG_SET_ORDER) != 0){
            $playurl = $body_links[0][1];
        }
        $data = [
            "title" => $json->desc,
            "desc" => $desc,
            "play" => $playurl
        ];

        return $data;
    }


    /**
     * 抖音内容抓取
     */
    public function catch($link)
    {
        header('Access-Control-Allow-Origin:*');
        header('Content-type:application/json; charset=utf-8');
        $url = $link;
        $ids  = explode('/', parse_url($this->getrealurl($url))['path'])[2];
        $json = json_decode($this->get_curl('https://www.iesdouyin.com/web/api/v2/aweme/iteminfo/?item_ids=' . $ids))->item_list[0];
        // 描述获取
        $desc = $json->share_info->share_weibo_desc;
        $play =  $this->get_curl(str_ireplace('playwm', 'play', $json->video->play_addr->url_list[0]), ['ua' => 'Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/75.0.3770.100 Mobile Safari/537.36']);
        preg_match_all("/<a[^<>]+href *= *[\"']?(http\:\/\/[^ '\"]+)/i", $play, $body_links, PREG_SET_ORDER);
        $data = [
            "title" => $json->desc,
            "desc" => $desc,
            "play" => $body_links[0][1]
        ];

        // 将视频文件下载
        $r = file_get_contents($data['play']);
        $filename = './upload' . DIRECTORY_SEPARATOR . uniqid() . '.mp4';
        file_put_contents($filename, $r);
        $data['video'] = ltrim($filename, '.');

        return json(['code' => 0, 'msg' => '获取成功', 'data' => $data]);
    }

    /**
     * 根据路径获取访问内容
     */
    private function get_curl($url, $paras = array())
    {
        $ch = curl_init();
        curl_setopt($ch, CURLOPT_URL, $url);
        curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false);
        curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, false);
        if (isset($paras['httpheader'])) { // 请求头
            $httpheader = $paras['httpheader'];
        } else {
            $httpheader[] = "Accept:*/*";
            $httpheader[] = "Accept-Encoding:gzip,deflate,sdch";
            $httpheader[] = "Accept-Language:zh-CN,zh;q=0.8";
            $httpheader[] = "Connection:close";
        }
        curl_setopt($ch, CURLOPT_HTTPHEADER, $httpheader);
        if (isset($paras['ctime'])) { // 连接超时
            curl_setopt($ch, CURLOPT_CONNECTTIMEOUT_MS, $paras['ctime']);
        }
        if (isset($paras['rtime'])) { // 读取超时
            curl_setopt($ch, CURLOPT_TIMEOUT_MS, $paras['rtime']);
        }
        if (isset($paras['post'])) {
            curl_setopt($ch, CURLOPT_POST, 1);
            curl_setopt($ch, CURLOPT_POSTFIELDS, $paras['post']);
        }
        if (isset($paras['header'])) {
            curl_setopt($ch, CURLOPT_HEADER, true);
        }
        if (isset($paras['cookie'])) {
            curl_setopt($ch, CURLOPT_COOKIE, $paras['cookie']);
        }
        if (isset($paras['refer'])) {
            if ($paras['refer'] == 1) {
                curl_setopt($ch, CURLOPT_REFERER, 'http://m.qzone.com/infocenter?g_f=');
            } else {
                curl_setopt($ch, CURLOPT_REFERER, $paras['refer']);
            }
        }
        if (isset($paras['ua'])) {
            curl_setopt($ch, CURLOPT_USERAGENT, $paras['ua']);
        } else {
            curl_setopt($ch, CURLOPT_USERAGENT, "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36");
        }
        if (isset($paras['nobody'])) {
            curl_setopt($ch, CURLOPT_NOBODY, 1);
        }
        curl_setopt($ch, CURLOPT_ENCODING, "gzip");
        curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
        $ret = curl_exec($ch);
        curl_close($ch);
        return $ret;
    }

    /**
     * 获取真是的访问地址
     */
    private function getrealurl($url)
    {
        $header = get_headers($url, 1);
        if (strpos($header[0], '301') || strpos($header[0], '302')) {
            if (is_array($header['Location'])) {
                return $header['Location'][count($header['Location']) - 1];
            } else {
                return $header['Location'];
            }
        } else {
            return $url;
        }
    }
}