python 爬虫爬爆米花视频仅供学习

直接上干货

一.页面分析

1.打开Chrome,输入 https://www.baomihua.com/

2.随便打开一个视频页

网址为：https://video.baomihua.com/v/37869538?reportto=REC_37869538

经测试：?reportto=REC_37869538 可以不加

所以入口地址为：https://video.baomihua.com/v/37869538

效果如下图：

打开开发者工具或按F12

按Alt+F5 刷新页面抓包

分析出视频的真正播放地址

在以上图片上可以看出

第1个media 的Status为(canceled) 所以忽略掉

第2个media 的Status为302, Type为空，这个重定向拿到了真实的播放地址，如下图

第3个media就是我们要的数据源

经以上数据分析得出：

最后的真实视频地址url：

http://219.135.240.146/vm36003.baomihua.com/dad0a3a539b1a623b61cf793948cd6ed/5B286BA1/3787/37869538_7_eaf7b02fbe74965404c0e7e301b2b3f7.mp4?wsrid_tag=5b286b03_PSgdhzdx7zz143_9439-19129&wsiphost=local

初步的视频地址url：

http://vm36003.baomihua.com/dad0a3a539b1a623b61cf793948cd6ed/5B286BA1/3787/37869538_7_eaf7b02fbe74965404c0e7e301b2b3f7.mp4

获得视频地址的url:

https://play.baomihua.com/getvideourl.aspx?jsoncallback=jQuery112409860452230116918_1529375493874&flvid=37869538&devicetype=pc_noflash&dataType=json&_=1529375493875

接着我们要做：先获得视频地址url,接着得到初步视频地址url,最后得到真实视频地址url

获得视频地址的url参数分析：

jsoncallback=jQuery112409860452230116918_1529375493874 经测试，jQuery后面的数字是随机的

flvid=37869538 传入视频id

devicetype=pc_noflash 不变

dataType=json 不变

_=1529375493875 不变

   获得视频地址的url请求头信息 
 

referer: https://video.baomihua.com/v/37869538

请求发出后得到数据

通过上图数据可以得到初步的视频地址url

初步的视频地址url地址分析：

http://vm36003.baomihua.com/dad0a3a539b1a623b61cf793948cd6ed/5B286BA1/3787/37869538_7_eaf7b02fbe74965404c0e7e301b2b3f7.mp4

对应

http://host/dir/stream_name.videofiletype

以下是python2.7代码

# coding:utf-8

import requests
import urllib
import re
import os
import json
import random
import time
from hashlib import md5
from requests.exceptions import RequestException

class BaomihuaMovie(object):

    #基本链接
    base_url = 'http://video.baomihua.com/v/'

    #请求头信息
    headers = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.62 Safari/537.36'
    }

    # 获取影片数据url
    # 参数1：{0}生成jQuery动态函数，{1}影片id
    playurl = 'http://play.baomihua.com/getvideourl.aspx?jsoncallback=jQuery{0}&flvid={1}&devicetype=pc_noflash&dataType=json'

    # 获取影片ids url
    # 参数1：{0}类型，{1}每页多少个，{2}当前在第几页
    movie_ids = 'http://www.baomihua.com/interfaces/getindexinfo.ashx?datatype=VideoListRec&typeid={0}&pagesize={1}&curpage={2}&scenetype=pc_channel'

    # 获取影片movie url1
    # 参数1：{0}jsondata['host']，{1}jsondata['dir']，{2}jsondata['stream_name']， {3}jsondata['videofiletype']
    movie_url1 = 'http://{0}/{1}/{2}.{3}'

    '''
    获取播放地址
    :param flvid: 影片id
    :return jsondata 影片数据
    '''
    def get_movie_data(self, flvid):
        url = self.playurl.format(random.randint(100000000000000, 999999999999999), flvid)
        print '第一个解释: ' + url
        self.headers['Host'] = 'vm36003.baomihua.com'
        self.headers['Referer'] = 'http://video.baomihua.com/v/' + str(flvid)
        response = requests.get(url, headers=self.headers)
        if response.status_code == 200:
            pattern = re.compile('.*?\((.*?)\).*?', re.S)
            result = re.search(pattern, response.content).group(1)
            jsondata = json.loads(result)
            return jsondata
        else:
            print 'get_play_data error'

    '''
    解释播放地址并保存到本地
    :param flvid 视频id
    '''
    def parse_video(self, flvid):
        jsondata = self.get_movie_data(flvid)
        if jsondata:
            filename = urllib.unquote(str(jsondata['title']));
            print '影片名：' + filename
            movieurl1 = self.movie_url1.format(jsondata['host'], jsondata['dir'], jsondata['stream_name'],jsondata['videofiletype'])
            print '第二个解释: ' + movieurl1

            self.headers['Host'] = 'vm36003.baomihua.com'
            self.headers['Referer'] = 'http://video.baomihua.com/v/' + str(id)
            movie = requests.get(movieurl1, headers=self.headers, allow_redirects=False)  #allow_redirects=False 为拒绝默认的301/302重定向从而可以通过html.headers[‘Location’]拿到重定向的URL
            if movie.status_code == 200:
                with open(filename.decode('utf-8') + '.' + jsondata['videofiletype'], 'wb') as mp4file:
                    mp4file.write(movie.content)
                print 'movie url1 parse_video ok'
            elif movie.status_code == 302:
                movieurl2 = movie.headers['Location']
                print '第三个解释: ' + movieurl2
                pattern = re.compile('.*?//(.*?)/.*?')
                movieip = re.match(pattern, movieurl2).group(1)
                # print movieip
                self.headers['Host'] = movieip
                movie2 = requests.get(movieurl2, headers=self.headers)
                if movie2.status_code == 200 or movie2.status_code == 206:
                    with open(filename.decode('utf-8') + '.' + jsondata['videofiletype'], 'wb') as mp4file:
                        mp4file.write(movie2.content)
                    print 'movie url2 parse_video ok'
            else:
                print 'parse_video error'
        else:
            print 'no jsondata'

    '''
    获取视频id组
    :param label    标签名
    :param typeid   标签名对应的类型id
    :param pagesize 一页共显示多个条记录
    :param curpage  当前第几页
    '''
    def get_movie_id(self, label, typeid, pagesize, curpage):
        url = self.movie_ids.format(typeid, pagesize, curpage)
        self.headers['Host'] = 'www.baomihua.com'
        self.headers['Referer'] = 'http://www.baomihua.com/' + label
        result = requests.get(url, headers = self.headers)
        if result.status_code == 200:
            resultjson = result.json()
            items = resultjson['Videolist']
            if items:
                for item in items:
                    yield item['videoId']
        else:
            print '链接解释错误！！！'

if __name__=='__main__':
    bmh = BaomihuaMovie()
    label = 'funny'     #搞笑
    typeid = 3          #搞笑类目对应的id号
    pagesize = 5
    curpage = 1
    #label 与typeid 要对应
    for flvid in bmh.get_movie_id(label, typeid, pagesize, curpage):
        print '开始解释： ' + str(flvid)
        bmh.parse_video(flvid)
        print '-' * 50
        time.sleep(random.randint(1,8))

    # bmh.parse_video(37873004)
    # url = 'http://play.baomihua.com/getvideourl.aspx?jsoncallback=jQuery1124032523956343&flvid=37797680&devicetype=pc_noflash&dataType=json&_=1529120814393'

python 爬虫 爬爆米花视频 仅供学习

猜你喜欢

python 爬虫爬爆米花视频仅供学习