python3 下載bilibili視頻 腳本

code

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


'''
項目: B站動漫番劇(bangumi)下載
版本2: 無加密API版,但是需要加入登錄后cookie中的SESSDATA字段,才可下載720p及以上視頻
API:
1.獲取cid的api為 https://api.bilibili.com/x/web-interface/view?aid=47476691 aid后面為av號
2.下載鏈接api為 https://api.bilibili.com/x/player/playurl?avid=44743619&cid=78328965&qn=32 cid為上面獲取到的 avid為輸入的av號 qn為視頻質量
注意:
但是此接口headers需要加上登錄后'Cookie': 'SESSDATA=3c5d20cf%2C1556704080%2C7dcd8c41' (30天的有效期)(因為現在只有登錄后才能看到720P以上視頻了)
不然下載之后都是最低清晰度,哪怕選擇了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

# 訪問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', # 登錄B站后復制一下cookie中的SESSDATA字段,有效期1個月
        'Host': 'api.bilibili.com'
    }
    html = requests.get(url_api, headers=headers).json()
    # print(html)
    # 當下載會員視頻時,如果cookie中傳入的不是大會員的SESSDATA時就會返回: {'code': -404, 'message': '啥都木有', 'ttl': 1, 'data': None}
    if html['code'] != 0:
        print('注意!當前集數為B站大會員專享,若想下載,Cookie中請傳入大會員的SESSDATA')
        return 'NoVIP'
    video_list = []
    for i in html['data']['durl']:
        video_list.append(i['url'])
    print(video_list)
    return video_list


# 下載視頻
'''
 urllib.urlretrieve 的回調函數：
def callbackfunc(blocknum, blocksize, totalsize):
    @blocknum:  已經下載的數據塊
    @blocksize: 數據塊的大小
    @totalsize: 遠程文件的大小
'''


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

    # 設置下載進度條
    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

    # 設置下載進度條
    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')


# 字節bytes轉化K\M\G
def format_size(bytes):
    try:
        bytes = float(bytes)
        kb = bytes / 1024
    except:
        print("傳入的字節格式不對")
        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)


#  下載視頻
def down_video(video_list, title, start_url, page):
    num = 1
    print('[正在下載第{}話視頻,請稍等...]:'.format(page) + title)
    currentVideoPath = os.path.join(sys.path[0], 'bilibili_video', title)  # 當前目錄作為下載目錄
    for i in video_list:
        opener = urllib.request.build_opener()
        # 請求頭
        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 的值要為 bytes=0- 才能下載完整視頻
            ('Referer', start_url),  # 注意修改referer,必須要加的!
            ('Origin', 'https://www.bilibili.com'),
            ('Connection', 'keep-alive'),

        ]
        urllib.request.install_opener(opener)
        # 創建文件夾存放下載的視頻
        if not os.path.exists(currentVideoPath):
            os.makedirs(currentVideoPath)
        # 開始下載
        if len(video_list) > 1:
            urllib.request.urlretrieve(url=i, filename=os.path.join(currentVideoPath, r'{}-{}.flv'.format(title, num)),
                                       reporthook=Schedule_cmd)  # 寫成mp4也行  title + '-' + num + '.flv'
        else:
            urllib.request.urlretrieve(url=i, filename=os.path.join(currentVideoPath, r'{}.flv'.format(title)),
                                       reporthook=Schedule_cmd)  # 寫成mp4也行  title + '-' + num + '.flv'
        num += 1


# 合並視頻(20190802新版)
def combine_video(title_list):
    video_path = os.path.join(sys.path[0], 'bilibili_video')  # 下載目錄
    for title in title_list:
        current_video_path = os.path.join(video_path ,title)
        if len(os.listdir(current_video_path)) >= 2:
            # 視頻大於一段才要合並
            print('[下載完成,正在合並視頻...]:' + title)
            # 定義一個數組
            L = []
            # 遍歷所有文件
            for file in sorted(os.listdir(current_video_path), key=lambda x: int(x[x.rindex("-") + 1:x.rindex(".")])):
                # 如果后綴名為 .mp4/.flv
                if os.path.splitext(file)[1] == '.flv':
                    # 拼接成完整路徑
                    filePath = os.path.join(current_video_path, file)
                    # 載入視頻
                    video = VideoFileClip(filePath)
                    # 添加到數組
                    L.append(video)
            # 拼接視頻
            final_clip = concatenate_videoclips(L)
            # 生成目標視頻文件
            final_clip.to_videofile(os.path.join(current_video_path, r'{}.mp4'.format(title)), fps=24, remove_temp=False)
            print('[視頻合並完成]' + title)
        else:
            # 視頻只有一段則直接打印下載完成
            print('[視頻合並完成]:' + title)


if __name__ == '__main__':
    start_time = time.time()
    # 用戶輸入番劇完整鏈接地址
    # 1. https://www.bilibili.com/bangumi/play/ep267692 (用帶ep鏈接)
    # 2. https://www.bilibili.com/bangumi/play/ss26878  (不要用這個ss鏈接,epinfo的aid會變成'-1')
    print('*' * 30 + 'B站番劇視頻下載小助手' + '*' * 30)
    print('[提示]: 1.如果您想下載720P60,1080p+,1080p60質量的視頻,請將35行代碼中的SESSDATA改成你登錄大會員后得到的SESSDATA,普通用戶的SESSDATA最多只能下載1080p的視頻')
    print('       2.若發現下載的視頻質量在720p以下,請將35行代碼中的SESSDATA改成你登錄后得到的SESSDATA(有效期一個月),而失效的SESSDATA就只能下載480p的視頻')

    start = input('請輸入您要下載的B站番劇的完整鏈接地址(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('您將要下載的番劇名為:' + ep_info['mediaInfo']['title']) # 字段格式太不統一了
    y = input('請輸入1或2 - 1.只下載當前一集 2.下載此番劇的全集:')
    # 1.如果只下載當前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']) + '話 ' + ep_info['videoData']['index_title']])
    # 2.下載此番劇全部ep
    else:
        for i in ep_info['epList']:
            # if i['badge'] == '': # 當badge字段為'會員'時,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']) + '話 ' + i['index_title']])

    # qn參數就是視頻清晰度
    # 可選值：
    # 116: 高清1080P60 (需要帶入大會員的cookie中的SESSDATA才行,普通用戶的SESSDATA最多只能下載1080p的視頻,不帶入SESSDATA就只能下載480p的)
    # 112: 高清1080P+ (hdflv2) (需要大會員)
    # 80: 高清1080P (flv)
    # 74: 高清720P60 (需要大會員)
    # 64: 高清720P (flv720)
    # 32: 清晰480P (flv480)
    # 16: 流暢360P (flv360)
    print('請輸入您要下載視頻的清晰度(1080p60:116;1080p+:112;1080p:80;720p60:74;720p:64;480p:32;360p:16; **注意:1080p+,1080p60,720p60都需要帶入大會員的cookie中的SESSDATA才行,普通用戶的SESSDATA最多只能下載1080p的視頻):')
    quality = input('請輸入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)  # 替換為空的
        print('[下載番劇標題]:' + 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)
        # 定義線程
        if video_list != 'NoVIP':
            th = threading.Thread(target=down_video, args=(video_list, title, start_url, page))
            # 將線程加入線程池
            threadpool.append(th)
        page += 1

    # 開始線程
    for th in threadpool:
        th.start()
    # 等待所有線程運行完畢
    for th in threadpool:
        th.join()
    
    # 最后合並視頻
    print(title_list)
    combine_video(title_list)
    
    end_time = time.time()  # 結束時間
    print('下載總耗時%.2f秒,約%.2f分鍾' % (end_time - start_time, int(end_time - start_time) / 60))
    # 如果是windows系統，下載完成后打開下載目錄
    currentVideoPath = os.path.join(sys.path[0], 'bilibili_video')  # 當前目錄作為下載目錄
    if (sys.platform.startswith('win')):
        os.startfile(currentVideoPath)