python 爬取腾讯视频弹幕

腾讯视频弹幕地址:http://mfm.video.qq.com/danmu?timestamp=0&target_id=xxxxx 

1.在视频地址中获取target_id

2.通过视频cid和vid获取target_id

3.cid和vid在视频链接网页中可以抓取

4.在电视剧任意一集页面中可以抓取每一集vid，cid一样。

5.弹幕的timestamp参数每隔30s变化一次

代码

import requests
import json
import random
import os
from lxml import etree
from requests.packages.urllib3.exceptions import InsecureRequestWarning
requests.packages.urllib3.disable_warnings(InsecureRequestWarning)


headers = {
    'USER-AGENT': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.99 Safari/537.36',
}

def get_str_btw(s, f, b):
    par = s.partition(f)
    return (par[2].partition(b))[0][:]

def filename_find(filepath, return_type=0):
    basename = os.path.basename(filepath)
    extension = f'.{basename.split(".")[-1]}'
    extension_lang = f'.{basename.split(".")[-2]}.{basename.split(".")[-1]}'
    if not basename.__contains__('.'):
        extension = ''
        extension_lang = ''
    filename_without_extension = basename[0:len(basename)-len(extension)]
    if return_type == 0:    # 文件名
        return basename
    if return_type == 1:    # 后缀名
        return extension
    if return_type == 2:    # 无后缀文件名
        return filename_without_extension
    if return_type == 3:    # 后缀名加字幕语言名
        return extension_lang


def get_targetid(cid,vid):  # 传入豆瓣api获得的腾讯视频uri
    base_url = 'https://access.video.qq.com/danmu_manage/regist?vappid=97767206&vsecret=c0bdcbae120669fff425d0ef853674614aa659c605a613a4'
    # print(get_targetid(url))
    # 构造请求头部
    headers = {
    'USER-AGENT':'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.99 Safari/537.36',
    }
    data = {"wRegistType":2,"vecIdList":[vid],"wSpeSource":0,"bIsGetUserCfg":1,"mapExtData":{vid:{"strCid":cid,"strLid":""}}}
    # 发送请求，获得响应
    response = requests.post(url=base_url, headers=headers, data=json.dumps(data))
    # 获得网页源代码
    html = response.text
    targetid = get_str_btw(html, 'targetid=', '&')
    # 返回targetid
    return targetid


def seconds_to_time(seconds):
    mm = int(seconds * 100 % 100)
    m, s = divmod(seconds, 60)
    h, m = divmod(m, 60)
    return ("%d:%02d:%02d.%02d" % (h, m, s, mm))


def sort_comment(timestamp,res):
    comments = []
    con = sorted(res["comments"], key=lambda keys: keys['timepoint'])
    for i in range(timestamp-15, timestamp+15):
        com = []
        for j in con:
            if j['timepoint'] == i:
                com.append(j)
        comments.append(com)
    return comments


def down_qq_danmu(targetid, danmu_name):
    url = 'https://mfm.video.qq.com/danmu?target_id=' + targetid
    try:
        os.remove(danmu_name)
    except:
        pass
    with open(danmu_name, "a", encoding='utf-8') as file:
        file.write('''[Script Info]
    Title: bilibili ASS 弹幕在线转换
    Original Script: 根据 251884753.xml 的弹幕信息，由 https://github.com/tiansh/us-danmaku 生成
    ScriptType: v4.00+
    Collisions: Normal
    PlayResX: 560
    PlayResY: 420
    Timer: 10.0000
    
    [V4+ Styles]
    Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
    Style: Fix,Microsoft YaHei UI,20,&H66FFFFFF,&H66FFFFFF,&H66000000,&H66000000,1,0,0,0,100,100,0,0,1,2,0,2,20,20,2,0
    Style: R2L,Microsoft YaHei UI,20,&H66FFFFFF,&H66FFFFFF,&H66000000,&H66000000,1,0,0,0,100,100,0,0,1,2,0,2,20,20,2,0
    
    [Events]
    Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
    ''')
        n = 0
        for i in range(15,100000,30):
            try:
                data = {
                    "timestamp":i
                }
                #print(i)
                response = requests.get(url,headers=headers,params=data,verify=False)
                res = json.loads(response.content)     # 字符串转化为列表格式
                if res['count'] != 0: #判断弹幕数量，确实是否爬取结束
                    comments = sort_comment(i,res)
                    for content in comments:
                        try:
                            ram = random.randint(0, len(content) - 1)   # 随机数
                            content_choose = content[ram]['content']  # 爱奇艺弹幕中每秒有多条 随机从中选取一条弹幕作为展示
                            seconds = content[ram]['timepoint']
                            line = [20, 40, 60, 80]
                            style = '\move(%s,%s,%s,%s)' % (random.randint(560, 760), line[n], -random.randint(10, 100), line[n])
                            # file.write('Dialogue: 0,%(start)s,%(end)s,%(styleid)s,,20,20,2,,{%(styles)s}%(text)s\n' % {'start': seconds_to_time(float(seconds)+ random.random()), 'end': seconds_to_time(float(seconds)+ random.random() + 8), 'styles': styles[i], 'text': item.getElementsByTagName('content')[0].firstChild.data, 'styleid': 'Fix'})
                            file.write('Dialogue: 0,%(start)s,%(end)s,%(styleid)s,,20,20,2,,{%(styles)s}%(text)s\n' % {
                                'start': seconds_to_time(float(seconds) + random.random()),
                                'end': seconds_to_time(float(seconds) + random.random() + 8), 'styles': style,
                                'text': content_choose,
                                'styleid': 'Fix'})
                            n = n + 1
                            if n > 3:
                                n = 0
                        except:
                            pass
                else:
                    print('爬完了')
                    break
            except:
                pass


def qqvideo_bullet(file_path, cid, vid):
    targetid = get_targetid(cid, vid)
    danmu_name = os.path.dirname(file_path) + "/" + filename_find(file_path, 2) + ".Danish.ass"
    down_qq_danmu(targetid, danmu_name)
    return danmu_name


def get_vid_list(url):      #根据任意一集的url获取电视剧所有集的vid
    res = requests.get(url).text
    html_elem = etree.HTML(res)
    info = html_elem.xpath('//html/head/script[@r-notemplate="true"][@type="text/javascript"]/text()')[0]
    # print(info)
    vid_info = json.loads(get_str_btw(info, '"vip_ids":', ',"comment_show_type"'))
    vid_list = []
    for vid in vid_info:
        vid_list.append(vid['V'])
    return vid_list


if __name__ == "__main__":
    uri = "tenvideo2://?action=1&cover_id=ccd36tevohjt9ze&video_id=u0035tjlfmu&from=douban.doc_m"   # 豆瓣返回腾讯视频uri
    cid = get_str_btw(uri, 'cover_id=', '&')
    vid = get_str_btw(uri, 'video_id=', '&')
    file_path = r'I:\python练习\爱奇艺弹幕下载转换\[超体]Lucy.2014.BluRay.720p.x264.AC3-BtsHD.mkv'
    #qqvideo_bullet(file_path, cid, vid)
    url = "https://v.qq.com/x/cover/mzc00200js3mdvw/q00354i139r.html"
    vid_list = get_vid_list(url)