利用python 下载bilibili视频

运行效果：

完整代码

# !/usr/bin/python
# -*- coding:utf-8 -*-
# time: 2019/07/21--20:12
__author__ = 'henry'


'''
项目: b站动漫番剧(bangumi)下载
版本2: 无加密api版,但是需要加入登录后cookie中的sessdata字段,才可下载720p及以上视频
api:
1.获取cid的api为 https://api.bilibili.com/x/web-interface/view?aid=47476691 aid后面为av号
2.下载链接api为 https://api.bilibili.com/x/player/playurl?avid=44743619&cid=78328965&qn=32 cid为上面获取到的 avid为输入的av号 qn为视频质量
注意:
但是此接口headers需要加上登录后'cookie': 'sessdata=3c5d20cf%2c1556704080%2c7dcd8c41' (30天的有效期)(因为现在只有登录后才能看到720p以上视频了)
不然下载之后都是最低清晰度,哪怕选择了80也是只有480p的分辨率!!
'''

import requests, time, urllib.request, re
from moviepy.editor import *
import os, sys, threading, json

import imageio

# 访问api地址
def get_play_list(aid, cid, quality):
  url_api = 'https://api.bilibili.com/x/player/playurl?cid={}&avid={}&qn={}'.format(cid, aid, quality)
  headers = {
    'user-agent': 'mozilla/5.0 (windows nt 6.1; wow64) applewebkit/537.36 (khtml, like gecko) chrome/55.0.2883.87 safari/537.36',
    'cookie': 'sessdata=13bd2abb%2c1619949439%2c2815d*b1', # 登录b站后复制一下cookie中的sessdata字段,有效期1个月
    'host': 'api.bilibili.com'
  }
  html = requests.get(url_api, headers=headers).json()
  # print(html)
  # 当下载会员视频时,如果cookie中传入的不是大会员的sessdata时就会返回: {'code': -404, 'message': '啥都木有', 'ttl': 1, 'data': none}
  if html['code'] != 0:
    print('注意!当前集数为b站大会员专享,若想下载,cookie中请传入大会员的sessdata')
    return 'novip'
  video_list = []
  for i in html['data']['durl']:
    video_list.append(i['url'])
  print(video_list)
  return video_list


# 下载视频
'''
 urllib.urlretrieve 的回调函数：
def callbackfunc(blocknum, blocksize, totalsize):
  @blocknum: 已经下载的数据块
  @blocksize: 数据块的大小
  @totalsize: 远程文件的大小
'''


def schedule_cmd(blocknum, blocksize, totalsize):
  speed = (blocknum * blocksize) / (time.time() - start_time)
  # speed_str = " speed: %.2f" % speed
  speed_str = " speed: %s" % format_size(speed)
  recv_size = blocknum * blocksize

  # 设置下载进度条
  f = sys.stdout
  pervent = recv_size / totalsize
  percent_str = "%.2f%%" % (pervent * 100)
  n = round(pervent * 50)
  s = ('#' * n).ljust(50, '-')
  f.write(percent_str.ljust(8, ' ') + '[' + s + ']' + speed_str)
  f.flush()
  # time.sleep(0.1)
  f.write('\r')


def schedule(blocknum, blocksize, totalsize):
  speed = (blocknum * blocksize) / (time.time() - start_time)
  # speed_str = " speed: %.2f" % speed
  speed_str = " speed: %s" % format_size(speed)
  recv_size = blocknum * blocksize

  # 设置下载进度条
  f = sys.stdout
  pervent = recv_size / totalsize
  percent_str = "%.2f%%" % (pervent * 100)
  n = round(pervent * 50)
  s = ('#' * n).ljust(50, '-')
  print(percent_str.ljust(6, ' ') + '-' + speed_str)
  f.flush()
  time.sleep(2)
  # print('\r')


# 字节bytes转化k\m\g
def format_size(bytes):
  try:
    bytes = float(bytes)
    kb = bytes / 1024
  except:
    print("传入的字节格式不对")
    return "error"
  if kb >= 1024:
    m = kb / 1024
    if m >= 1024:
      g = m / 1024
      return "%.3fg" % (g)
    else:
      return "%.3fm" % (m)
  else:
    return "%.3fk" % (kb)


# 下载视频
def down_video(video_list, title, start_url, page):
  num = 1
  print('[正在下载第{}话视频,请稍等...]:'.format(page) + title)
  currentvideopath = os.path.join(sys.path[0], 'bilibili_video', title) # 当前目录作为下载目录
  for i in video_list:
    opener = urllib.request.build_opener()
    # 请求头
    opener.addheaders = [
      # ('host', 'upos-hz-mirrorks3.acgvideo.com'), #注意修改host,不用也行
      ('user-agent', 'mozilla/5.0 (macintosh; intel mac os x 10.13; rv:56.0) gecko/20100101 firefox/56.0'),
      ('accept', '*/*'),
      ('accept-language', 'en-us,en;q=0.5'),
      ('accept-encoding', 'gzip, deflate, br'),
      ('range', 'bytes=0-'), # range 的值要为 bytes=0- 才能下载完整视频
      ('referer', start_url), # 注意修改referer,必须要加的!
      ('origin', 'https://www.bilibili.com'),
      ('connection', 'keep-alive'),

    ]
    urllib.request.install_opener(opener)
    # 创建文件夹存放下载的视频
    if not os.path.exists(currentvideopath):
      os.makedirs(currentvideopath)
    # 开始下载
    if len(video_list) > 1:
      urllib.request.urlretrieve(url=i, filename=os.path.join(currentvideopath, r'{}-{}.flv'.format(title, num)),
                    reporthook=schedule_cmd) # 写成mp4也行 title + '-' + num + '.flv'
    else:
      urllib.request.urlretrieve(url=i, filename=os.path.join(currentvideopath, r'{}.flv'.format(title)),
                    reporthook=schedule_cmd) # 写成mp4也行 title + '-' + num + '.flv'
    num += 1


# 合并视频(20190802新版)
def combine_video(title_list):
  video_path = os.path.join(sys.path[0], 'bilibili_video') # 下载目录
  for title in title_list:
    current_video_path = os.path.join(video_path ,title)
    if len(os.listdir(current_video_path)) >= 2:
      # 视频大于一段才要合并
      print('[下载完成,正在合并视频...]:' + title)
      # 定义一个数组
      l = []
      # 遍历所有文件
      for file in sorted(os.listdir(current_video_path), key=lambda x: int(x[x.rindex("-") + 1:x.rindex(".")])):
        # 如果后缀名为 .mp4/.flv
        if os.path.splitext(file)[1] == '.flv':
          # 拼接成完整路径
          filepath = os.path.join(current_video_path, file)
          # 载入视频
          video = videofileclip(filepath)
          # 添加到数组
          l.append(video)
      # 拼接视频
      final_clip = concatenate_videoclips(l)
      # 生成目标视频文件
      final_clip.to_videofile(os.path.join(current_video_path, r'{}.mp4'.format(title)), fps=24, remove_temp=false)
      print('[视频合并完成]' + title)
    else:
      # 视频只有一段则直接打印下载完成
      print('[视频合并完成]:' + title)


if __name__ == '__main__':
  start_time = time.time()
  # 用户输入番剧完整链接地址
  # 1. https://www.bilibili.com/bangumi/play/ep267692 (用带ep链接)
  # 2. https://www.bilibili.com/bangumi/play/ss26878 (不要用这个ss链接,epinfo的aid会变成'-1')
  print('*' * 30 + 'b站番剧视频下载小助手' + '*' * 30)
  print('[提示]: 1.如果您想下载720p60,1080p+,1080p60质量的视频,请将35行代码中的sessdata改成你登录大会员后得到的sessdata,普通用户的sessdata最多只能下载1080p的视频')
  print('    2.若发现下载的视频质量在720p以下,请将35行代码中的sessdata改成你登录后得到的sessdata(有效期一个月),而失效的sessdata就只能下载480p的视频')

  start = input('请输入您要下载的b站番剧的完整链接地址(例如:https://www.bilibili.com/bangumi/play/ep267692):')
  ep_url = start
  headers = {
    'user-agent': 'mozilla/5.0 (windows nt 6.1; wow64) applewebkit/537.36 (khtml, like gecko) chrome/55.0.2883.87 safari/537.36'
  }
  html = requests.get(ep_url,headers=headers).text
  ep_info = re.search(r'initial_state__=(.*?"]});', html).group(1)
  # print(ep_info)
  ep_info = json.loads(ep_info)
  print(ep_info)
  # print('您将要下载的番剧名为:' + ep_info['mediainfo']['title']) # 字段格式太不统一了
  y = input('请输入1或2 - 1.只下载当前一集 2.下载此番剧的全集:')
  # 1.如果只下载当前ep
  id_list = []
  if y == '1':
    try:
      id_list.append([ep_info['videodata']['aid'], ep_info['videodata']['cid'],
              ep_info['videodata']['title'] + ' ' + ep_info['videodata']['title']])
    except:
      id_list.append([ep_info['videodata']['aid'], ep_info['videodata']['cid'],
              '第' + str(ep_info['videodata']['index']) + '话 ' + ep_info['videodata']['index_title']])
  # 2.下载此番剧全部ep
  else:
    for i in ep_info['eplist']:
      # if i['badge'] == '': # 当badge字段为'会员'时,接口返回404
      try:
        id_list.append([i['aid'], i['cid'],
                i['titleformat'] + ' ' + i['title']])
      except:
        id_list.append([i['aid'], i['cid'],'第' + str(i['index']) + '话 ' + i['index_title']])

  # qn参数就是视频清晰度
  # 可选值：
  # 116: 高清1080p60 (需要带入大会员的cookie中的sessdata才行,普通用户的sessdata最多只能下载1080p的视频,不带入sessdata就只能下载480p的)
  # 112: 高清1080p+ (hdflv2) (需要大会员)
  # 80: 高清1080p (flv)
  # 74: 高清720p60 (需要大会员)
  # 64: 高清720p (flv720)
  # 32: 清晰480p (flv480)
  # 16: 流畅360p (flv360)
  print('请输入您要下载视频的清晰度(1080p60:116;1080p+:112;1080p:80;720p60:74;720p:64;480p:32;360p:16; **注意:1080p+,1080p60,720p60都需要带入大会员的cookie中的sessdata才行,普通用户的sessdata最多只能下载1080p的视频):')
  quality = input('请输入116或112或80或74或64或32或16:')
  threadpool = []
  title_list = []
  page = 1
  print(id_list)
  for item in id_list:
    aid = str(item[0])
    cid = str(item[1])
    title = item[2]
    title = re.sub(r'[\/\\:*?"<>|]', '', title) # 替换为空的
    print('[下载番剧标题]:' + title)
    title_list.append(title)
    start_url = ep_url
    video_list = get_play_list(aid, cid, quality)
    start_time = time.time()
    # down_video(video_list, title, start_url, page)
    # 定义线程
    if video_list != 'novip':
      th = threading.thread(target=down_video, args=(video_list, title, start_url, page))
      # 将线程加入线程池
      threadpool.append(th)
    page += 1

  # 开始线程
  for th in threadpool:
    th.start()
  # 等待所有线程运行完毕
  for th in threadpool:
    th.join()
  
  # 最后合并视频
  print(title_list)
  combine_video(title_list)
  
  end_time = time.time() # 结束时间
  print('下载总耗时%.2f秒,约%.2f分钟' % (end_time - start_time, int(end_time - start_time) / 60))
  # 如果是windows系统，下载完成后打开下载目录
  currentvideopath = os.path.join(sys.path[0], 'bilibili_video') # 当前目录作为下载目录
  if (sys.platform.startswith('win')):
    os.startfile(currentvideopath)

以上就是利用python 下载bilibili视频的详细内容，更多关于python 下载bilibili视频的资料请关注其它相关文章！