Merge branch 'feature/video_download' of lyq/crawl-bilibili into master

.gitignore

@@ -3,3 +3,5 @@
 /data
 /.vscode
 /utils/__pycache__/*.pyc
+/video_download/ffmpeg.linux64
+/video_download/bilibili_video/

video_download/.dockerignore

@@ -0,0 +1,13 @@
+node_modules
+npm-debug.log
+Dockerfile*
+docker-compose*
+.dockerignore
+.git
+.gitignore
+.env
+*/bin
+*/obj
+README.md
+LICENSE
+.vscode

video_download/.gitignore

@@ -0,0 +1 @@
+bilibili_video/

video_download/Dockerfile

@@ -0,0 +1,13 @@
+FROM python:3.6
+
+LABEL Name=bilibili_video_download Version=1.0.1
+EXPOSE 3000
+WORKDIR /app
+
+RUN mkdir -p ~/.pip
+ADD pip.conf /root/.pip/
+ADD requirements.txt /app/
+RUN python3 -m pip install -r requirements.txt
+ADD ffmpeg.linux64 /root/.imageio/ffmpeg/
+# ADD . /app
+CMD ["/bin/bash" ,"-c" ,"while true;do echo hello docker;sleep 1;done"]

video_download/README.md

@@ -0,0 +1,61 @@
+Bilibili(b站)视频下载 ![enter image description here](Pic/logo.png)
+===========================
+![](https://img.shields.io/badge/Python-3.6.3-green.svg) ![](https://img.shields.io/badge/requests-2.18.4-green.svg) ![](https://img.shields.io/badge/moviepy-0.2.3.2-green.svg)
+### Bilibili官网 - https://www.bilibili.com/
+|Author|:sunglasses:Henryhaohao:sunglasses:|
+|---|---
+|Email|:hearts:1073064953@qq.com:hearts:
+
+    
+****
+## :dolphin:声明
+### 软件均仅用于学习交流，请勿用于任何商业用途！感谢大家！
+## :dolphin:介绍
+### 该项目为[Bilibili(b站)](https://www.bilibili.com/)视频下载(支持分P多段视频的下载!)
+- **对于单P视频:直接传入B站av号或者视频链接地址(eg: 49842011或者https://www.bilibili.com/video/av49842011)**
+- **对于多P视频:**
+  > **1.下载全集:直接传入B站av号或者视频链接地址(eg: 49842011或者https://www.bilibili.com/video/av49842011)**<br>
+  > **2.下载其中一集:传入那一集的视频链接地址(eg: https://www.bilibili.com/video/av19516333/?p=2)**
+## :dolphin:爬虫版本
+- **版本一: bilibili_video_download_v1.py**
+  > **加密API版,不需要加入cookie,直接即可下载1080p视频<br>**
+- **版本二: bilibili_video_download_v2.py**
+  > **1.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**<br>
+  > **2.如果想下载1080p+视频,需要带入B站大会员的cookie中的SESSDATA才行,普通用户的SESSDATA最多只能下载1080p视频；请定期更换代码31行cookie中的SESSDATA值。跟换方法为：浏览器登录B站，在开发者模式(按F12) --> application --> cookie中找到SESSDATA值替换即可，一个月的有效期**
+- **版本三: bilibili_video_download_v3.py**
+  > **即版本二的升级版,为Threading多线程下载版本,下载速度大幅提升!**<br>  
+## :dolphin:运行环境
+Version: Python3
+## :dolphin:安装依赖库
+```
+pip3 install -r requirements.txt
+
+```
+## :dolphin:运行截图
+> - **运行下载**<br><br>
+![enter image description here](Pic/run.png)
+> - **下载完成**<br><br>
+![enter image description here](Pic/video.png)
+## :dolphin:**总结**
+> **最后，如果你觉得这个项目不错或者对你有帮助，给个Star呗，也算是对我学习路上的一种鼓励！<br>
+ 哈哈哈，感谢大家！笔芯哟~**:cupid::cupid:
+
+
+## docker运行
+
+windows中安装docker
+
+git clone https://xxx 本项目
+
+wget https://github.com/imageio/imageio-binaries/raw/master/ffmpeg/ffmpeg.linux64
+
+wget http://cdn.yoqi.me/direct/2019-07-09/ffmpeg.linux64
+
+docker build -t bilibili_video_download:latest .
+
+docker-compose up -d
+
+docker exec -it a93ac1 /bin/sh -c "[ -e /bin/bash ] && /bin/bash || /bin/sh"
+
+python3 bilibili_video_download_v1.py
+

video_download/bilibili_video_download_v1.py

@@ -0,0 +1,220 @@
+# !/usr/bin/python
+# -*- coding:utf-8 -*-
+# time: 2019/04/17--08:12
+__author__ = 'Henry'
+
+
+'''
+项目: B站视频下载
+
+版本1: 加密API版,不需要加入cookie,直接即可下载1080p视频
+
+20190422 - 增加多P视频单独下载其中一集的功能
+'''
+import imageio
+imageio.plugins.ffmpeg.download()
+
+import requests, time, hashlib, urllib.request, re, json
+from moviepy.editor import *
+import os, sys
+
+
+# 访问API地址
+def get_play_list(start_url, cid, quality):
+    entropy = 'rbMCKn@KuamXWlPMoJGsKcbiJKUfkPF_8dABscJntvqhRSETg'
+    appkey, sec = ''.join([chr(ord(i) + 2) for i in entropy[::-1]]).split(':')
+    params = 'appkey=%s&cid=%s&otype=json&qn=%s&quality=%s&type=' % (appkey, cid, quality, quality)
+    chksum = hashlib.md5(bytes(params + sec, 'utf8')).hexdigest()
+    url_api = 'https://interface.bilibili.com/v2/playurl?%s&sign=%s' % (params, chksum)
+    headers = {
+        'Referer': start_url,  # 注意加上referer
+        'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/55.0.2883.87 Safari/537.36'
+    }
+    # print(url_api)
+    html = requests.get(url_api, headers=headers).json()
+    # print(json.dumps(html))
+    video_list = [html['durl'][0]['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P{}段视频,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
+def combine_video(video_list, title):
+    currentVideoPath = os.path.join(sys.path[0], 'bilibili_video', title)  # 当前目录作为下载目录
+    if len(video_list) >= 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 = []
+        # 访问 video 文件夹 (假设视频都放在这里面)
+        root_dir = currentVideoPath
+        # 遍历所有文件
+        for file in sorted(os.listdir(root_dir), 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(root_dir, 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(root_dir, 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__':
+    # 用户输入av号或者视频链接地址
+    print('*' * 30 + 'B站视频下载小助手' + '*' * 30)
+    start = input('请输入您要下载的B站av号或者视频链接地址:')
+    if start.isdigit() == True:  # 如果输入的是av号
+        # 获取cid的api, 传入aid即可
+        start_url = 'https://api.bilibili.com/x/web-interface/view?aid=' + start
+    else:
+        # https://www.bilibili.com/video/av46958874/?spm_id_from=333.334.b_63686965665f7265636f6d6d656e64.16
+        start_url = 'https://api.bilibili.com/x/web-interface/view?aid=' + re.search(r'/av(\d+)/*', start).group(1)
+
+    # 视频质量
+    # <accept_format><![CDATA[flv,flv720,flv480,flv360]]></accept_format>
+    # <accept_description><![CDATA[高清 1080P,高清 720P,清晰 480P,流畅 360P]]></accept_description>
+    # <accept_quality><![CDATA[80,64,32,16]]></accept_quality>
+    quality = input('请输入您要下载视频的清晰度(1080p:80;720p:64;480p:32;360p:16)(填写80或64或32或16):')
+    # 获取视频的cid,title
+    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(start_url, headers=headers).json()
+    data = html['data']
+    video_title=data["title"].replace(" ","_")
+    cid_list = []
+    if '?p=' in start:
+        # 单独下载分P视频中的一集
+        p = re.search(r'\?p=(\d+)',start).group(1)
+        cid_list.append(data['pages'][int(p) - 1])
+    else:
+        # 如果p不存在就是全集下载
+        cid_list = data['pages']
+    # print(cid_list)
+    for item in cid_list:
+        cid = str(item['cid'])
+        title = item['part']
+        if not title:
+            title = video_title
+        title = re.sub(r'[\/\\:*?"<>|]', '', title)  # 替换为空的
+        print('[下载视频的cid]:' + cid)
+        print('[下载视频的标题]:' + title)
+        page = str(item['page'])
+        start_url = start_url + "/?p=" + page
+        video_list = get_play_list(start_url, cid, quality)
+        start_time = time.time()
+        down_video(video_list, title, start_url, page)
+        combine_video(video_list, title)
+
+    # 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)
+
+
+# 分P视频下载测试: https://www.bilibili.com/video/av19516333/

video_download/bilibili_video_download_v2.py

@@ -0,0 +1,233 @@
+# !/usr/bin/python
+# -*- coding:utf-8 -*-
+# time: 2019/04/16--17:12
+__author__ = 'Henry'
+
+
+'''
+项目: B站视频下载
+
+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!!
+
+20190422 - 增加多P视频单独下载其中一集的功能
+'''
+
+import requests, time, urllib.request, re
+from moviepy.editor import *
+import os, sys
+
+import imageio
+imageio.plugins.ffmpeg.download()
+
+# 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=aa15d6af%2C1560734457%2Ccc8ca251', # 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()
+    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P{}段视频,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
+def combine_video(video_list, title):
+    currentVideoPath = os.path.join(sys.path[0], 'bilibili_video', title)  # 当前目录作为下载目录
+    if len(video_list) >= 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 = []
+        # 访问 video 文件夹 (假设视频都放在这里面)
+        root_dir = currentVideoPath
+        # 遍历所有文件
+        for file in sorted(os.listdir(root_dir), 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(root_dir, 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(root_dir, 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__':
+    # 用户输入av号或者视频链接地址
+    print('*' * 30 + 'B站视频下载小助手' + '*' * 30)
+    start = input('请输入您要下载的B站av号或者视频链接地址:')
+    if start.isdigit() == True:
+        # 如果输入的是av号
+        # 获取cid的api, 传入aid即可
+        aid = start
+        start_url = 'https://api.bilibili.com/x/web-interface/view?aid=' + aid
+    else:
+        # 如果输入的是url (eg: https://www.bilibili.com/video/av46958874/)
+        aid = re.search(r'/av(\d+)/*', start).group(1)
+        start_url = 'https://api.bilibili.com/x/web-interface/view?aid=' + aid
+    # 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)
+    # 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+,1080p60,720p60,720p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('请填写116或112或80或74或64或32或16:')
+    # 获取视频的cid,title
+    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(start_url, headers=headers).json()
+    data = html['data']
+    cid_list = []
+    if '?p=' in start:
+        # 单独下载分P视频中的一集
+        p = re.search(r'\?p=(\d+)', start).group(1)
+        cid_list.append(data['pages'][int(p) - 1])
+    else:
+        # 如果p不存在就是全集下载
+        cid_list = data['pages']
+    # print(cid_list)
+    for item in cid_list:
+        cid = str(item['cid'])
+        title = item['part']
+        title = re.sub(r'[\/\\:*?"<>|]', '', title)  # 替换为空的
+        print('[下载视频的cid]:' + cid)
+        print('[下载视频的标题]:' + title)
+        page = str(item['page'])
+        start_url = start_url + "/?p=" + page
+        video_list = get_play_list(aid, cid, quality)
+        start_time = time.time()
+        down_video(video_list, title, start_url, page)
+        combine_video(video_list, title)
+
+    # 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)
+
+# 分P视频下载测试: https://www.bilibili.com/video/av19516333/

video_download/bilibili_video_download_v3.py

@@ -0,0 +1,235 @@
+# !/usr/bin/python
+# -*- coding:utf-8 -*-
+# time: 2019/07/02--08:12
+__author__ = 'Henry'
+
+
+'''
+项目: B站视频下载 - 多线程下载
+
+版本1: 加密API版,不需要加入cookie,直接即可下载1080p视频
+
+20190422 - 增加多P视频单独下载其中一集的功能
+20190702 - 增加视频多线程下载 速度大幅提升
+'''
+
+import requests, time, hashlib, urllib.request, re, json
+from moviepy.editor import *
+import os, sys, threading
+
+import imageio
+imageio.plugins.ffmpeg.download()
+
+# 访问API地址
+def get_play_list(start_url, cid, quality):
+    entropy = 'rbMCKn@KuamXWlPMoJGsKcbiJKUfkPF_8dABscJntvqhRSETg'
+    appkey, sec = ''.join([chr(ord(i) + 2) for i in entropy[::-1]]).split(':')
+    params = 'appkey=%s&cid=%s&otype=json&qn=%s&quality=%s&type=' % (appkey, cid, quality, quality)
+    chksum = hashlib.md5(bytes(params + sec, 'utf8')).hexdigest()
+    url_api = 'https://interface.bilibili.com/v2/playurl?%s&sign=%s' % (params, chksum)
+    headers = {
+        'Referer': start_url,  # 注意加上referer
+        'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/55.0.2883.87 Safari/537.36'
+    }
+    # print(url_api)
+    html = requests.get(url_api, headers=headers).json()
+    # print(json.dumps(html))
+    video_list = [html['durl'][0]['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P{}段视频,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
+def combine_video(video_list, title):
+    currentVideoPath = os.path.join(sys.path[0], 'bilibili_video', title)  # 当前目录作为下载目录
+    if len(video_list) >= 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 = []
+        # 访问 video 文件夹 (假设视频都放在这里面)
+        root_dir = currentVideoPath
+        # 遍历所有文件
+        for file in sorted(os.listdir(root_dir), 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(root_dir, 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(root_dir, 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av号或者视频链接地址
+    print('*' * 30 + 'B站视频下载小助手' + '*' * 30)
+    start = input('请输入您要下载的B站av号或者视频链接地址:')
+    if start.isdigit() == True:  # 如果输入的是av号
+        # 获取cid的api, 传入aid即可
+        start_url = 'https://api.bilibili.com/x/web-interface/view?aid=' + start
+    else:
+        # https://www.bilibili.com/video/av46958874/?spm_id_from=333.334.b_63686965665f7265636f6d6d656e64.16
+        start_url = 'https://api.bilibili.com/x/web-interface/view?aid=' + re.search(r'/av(\d+)/*', start).group(1)
+
+    # 视频质量
+    # <accept_format><![CDATA[flv,flv720,flv480,flv360]]></accept_format>
+    # <accept_description><![CDATA[高清 1080P,高清 720P,清晰 480P,流畅 360P]]></accept_description>
+    # <accept_quality><![CDATA[80,64,32,16]]></accept_quality>
+    quality = input('请输入您要下载视频的清晰度(1080p:80;720p:64;480p:32;360p:16)(填写80或64或32或16):')
+    # 获取视频的cid,title
+    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(start_url, headers=headers).json()
+    data = html['data']
+    cid_list = []
+    if '?p=' in start:
+        # 单独下载分P视频中的一集
+        p = re.search(r'\?p=(\d+)',start).group(1)
+        cid_list.append(data['pages'][int(p) - 1])
+    else:
+        # 如果p不存在就是全集下载
+        cid_list = data['pages']
+    # print(cid_list)
+    # 创建线程池
+    threadpool = []
+    for item in cid_list:
+        cid = str(item['cid'])
+        title = item['part']
+        title = re.sub(r'[\/\\:*?"<>|]', '', title)  # 替换为空的
+        print('[下载视频的cid]:' + cid)
+        print('[下载视频的标题]:' + title)
+        page = str(item['page'])
+        start_url = start_url + "/?p=" + page
+        video_list = get_play_list(start_url, cid, quality)
+        start_time = time.time()
+        # down_video(video_list, title, start_url, page)
+        # 定义线程
+        th = threading.Thread(target=down_video, args=(video_list, title, start_url, page))
+        # 将线程加入线程池
+        threadpool.append(th)
+        combine_video(video_list, title)
+
+    # 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()
+
+    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)
+
+
+# 分P视频下载测试: https://www.bilibili.com/video/av19516333/
+# 下载总耗时14.21秒,约0.23分钟

video_download/docker-compose.debug.yml

@@ -0,0 +1,10 @@
+version: '2.1'
+
+services:
+  bilibili_video_download:
+    image: bilibili_video_download
+    build:
+      context: .
+      dockerfile: Dockerfile
+    ports:
+      - 3000:3000

video_download/docker-compose.yml

@@ -0,0 +1,10 @@
+version: '2.1'
+
+services:
+  bilibili_video_download:
+    image: bilibili_video_download
+    build: .
+    ports:
+      - 3000:3000
+    volumes:
+      - "./:/app"

video_download/pip.conf

@@ -0,0 +1,3 @@
+[global]
+trusted-host=mirrors.aliyun.com
+index-url=http://mirrors.aliyun.com/pypi/simple/

video_download/requirements.txt

@@ -0,0 +1,3 @@
+moviepy==0.2.3.2
+requests==2.18.4
+scikit-image