美拍视频采集之视频地址解析下载

以此URl举个例子http://www.meipai.com/media/816057957

• 页面分析
  通过分析美拍播放页面的源代码，可以看到，播放页面的html代码中的有一个mate标签(property属性值为og:video:url)，该值就是页面的视频链接，只不过是经过加密后的

• 使用Java语言来自动获取美拍真实加密过的url地址，Crawl类

// 根据url请求html页面
public static String clawer2(String myurl) throws Exception{
        URL urlmy = new URL(myurl);
        HttpURLConnection con = (HttpURLConnection) urlmy.openConnection();
        con.setFollowRedirects(true);
        con.setInstanceFollowRedirects(false);
        con.connect();
        BufferedReader br = new BufferedReader(new InputStreamReader(con.getInputStream(),"UTF-8"));
        String s = "";
        StringBuffer sb = new StringBuffer("");
        while ((s = br.readLine()) != null) {
            sb.append(s+"\r\n");
        }
        return sb.toString();
    }

// 通过Jsoup包来解析html页面
public static String parsehtml(String url) throws Exception {
        String htmlContent = Crawl.clawer2(url);
        //使用jSoup解析里头的内容
        //就像操作html doc文档对象一样操作网页中的元素
        Document doc = Jsoup.parse(htmlContent);
        Element span = doc.select("head > meta:nth-child(16)").first();
        System.out.println(span.attr("content"));
        return span.attr("content");// 获取加密后的视频url地址
    }

• 解密加密字符串，解密后即可得到视频真实URL地址

/**
 * 解密加密过的字符串
 * Created by lihan on 2017/7/29.
 */
public class ParseUrl {
    private Map<String, String> dict1;
    private Map<String, String[]> dict2;
    
    // 获得16进制数，该数用来分割字符串
    public Map<String,String> getHex(String param1){
        dict1 = new HashMap<String, String>();
        String cstr =  param1.substring(4);//str
        String[] splitStr = param1.substring(0,4).split("");
        String hex = "";
        for (int i=3; i >= 0; i--){
            hex = hex + splitStr[i];
        }
        dict1.put("str", cstr);
        dict1.put("hex", hex);
        return dict1;
    }
    // 获取正确的字符串,解析16进制数
    public Map<String, String[]> getDecimal(String param1){
        dict2 = new HashMap<String, String[]>();
        // loc2是用来分割字符串的索引标识，转换16进制
        String loc2 = String.valueOf(Integer.parseInt(param1,16));
        String[] pre = loc2.substring(0,2).split("");//dict1.put("loc2", loc2.substring(0,2));
        String[] tail = loc2.substring(2).split("");
        dict2.put("pre", pre);
        dict2.put("tail", tail);
        return dict2;
    }
    // 分割字符串
    public String substr(String param1, String[] param2) {
        String loc3 = param1.substring(0, Integer.parseInt(param2[0]));//param2 = pu.getDec(pa2).get("pre")
        String loc4 = param1.substring(Integer.parseInt(param2[0]), Integer.parseInt(param2[0])+Integer.parseInt(param2[1]));
        return loc3 + param1.substring(Integer.parseInt(param2[0])).replace(loc4, "");
    }
    // 获取分割的位置
    public String[] getPosition(String param1, String[] param2){
        param2[0] = String.valueOf(param1.length() - Integer.parseInt(param2[0]) - Integer.parseInt(param2[1]));
        return param2;
    }

    // 程序入口
    public static void main(String[] args) throws Exception {
        ParseUrl pu = new ParseUrl();
        // 获取html中的加密字符串
        String code = Crawl.parsehtml("http://www.meipai.com/media/816057957");
        Map<String, String> dict2 = pu.getHex(code);
        Map<String,String[]> dict3 = pu.getDecimal(dict2.get("hex"));
        String str4 = pu.substr(dict2.get("str"), dict3.get("pre"));
        BASE64Decoder base64 = new BASE64Decoder();
        byte[] url = base64.decodeBuffer(pu.substr(str4, pu.getPosition(str4, dict3.get("tail"))));
        // 视频真实的url地址
        System.out.println(new String(url));
    }
}

• 运行后即可得到真实的视频url

• python版本可以参考：

需要安装requests库，安装命令:
pip install requests

# coding=utf-8
# Created by lihan on 2017/7/29.
import base64
import requests
from pyquery import PyQuery as pq

class Decode:
    def getHex(self,param1):
        return {
            'str': param1[4:],
            'hex': ''.join(list(param1[:4])[::-1]),
        }

    def getDecimal(self, param1):
        loc2 = str(int(param1, 16))
        print(loc2)#
        return {
            'pre': list(loc2[:2]),
            'tail': list(loc2[2:]),
        }

    def substr(self, param1, param2):
        loc3 = param1[0: int(param2[0])]
        loc4 = param1[int(param2[0]): int(param2[0]) + int(param2[1])]
        print("loc4",loc4)
        return loc3 + param1[int(param2[0]):].replace(loc4, "")

    def getPosition(self,param1, param2):
        param2[0] = len(param1) - int(param2[0]) - int(param2[1])
        return param2

    def decode(self, code):
        dict2 = self.getHex(code)
        dict3 = self.getDecimal(dict2['hex'])
        print("dict3", dict3['pre'])  #
        str4 = self.substr(dict2['str'], dict3['pre'])
        return base64.b64decode(self.substr(str4, self.getPosition(str4, dict3['tail'])))

    def crawl_video_url(self, url):
        response = requests.get(url).content
        d = pq(response)
        code= d('meta[property="og:video:url"]').attr('content')
        result = self.decode(code)
        print(result)


if __name__ == '__main__':
    url = "http://www.meipai.com/media/816057957"//任意播放页面的url
    Decode().crawl_video_url(url)