python爬取bilibili，下载视频-Toy模板网

这篇具有很好参考价值的文章主要介绍了python爬取bilibili，下载视频。希望对大家有所帮助。如果存在错误或未考虑完全的地方，请大家不吝赐教，您也可以点击"举报违法"按钮提交疑问。

一. 内容简介

python下载bilibili视频合集

二. 软件环境

2.1vsCode

2.2Anaconda

version: conda 22.9.0

2.3代码

链接：https://pan.baidu.com/s/1WuXTso_iltLlnrLffi1kYQ?pwd=1234

三.主要流程

3.1 下载单个视频

感觉现在下载的清晰度不够，可以在找找，给这块替换了就行
代码

import requests
import os
from lxml import etree
import re

def videoDownload1(url_):
    # 设置用户代理,cookie
    headers_ = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36',
        'Cookie': "buvid3=7014DDC0-BF1E-B121-F5A5-F10753C840B423630infoc; i-wanna-go-back=-1; _uuid=49BF2138-1E10F-D5F5-10898-D8311651B53927883infoc; FEED_LIVE_VERSION=V8; DedeUserID=171300042; DedeUserID__ckMd5=c65bec3211413192; CURRENT_FNVAL=4048; rpdid=|(J|)J~m~llk0J'uYm|)~klRl; header_theme_version=CLOSE; hit-new-style-dyn=1; hit-dyn-v2=1; is-2022-channel=1; fingerprint=fe5c7462625770aa2abce449a7c01fd2; buvid_fp_plain=undefined; b_nut=1691207170; b_ut=5; buvid_fp=fe5c7462625770aa2abce449a7c01fd2; LIVE_BUVID=AUTO4016915564967297; buvid4=1AE73807-AEA0-7078-DA57-7F9FE5C3D6F896987-023080912-A0g5nInZwV3VmJJT68FJxw%3D%3D; home_feed_column=5; SESSDATA=fc1266d3%2C1708653865%2C29c08%2A81-i-T9HQrucvpCVcPwSwXl5LmjTyduIzF9veu0KS9i2IwXK_xkcqlt1XQyxJ3sG-9HMSwLwAAKgA; bili_jct=068bc0a79f3fa7aa1a030e478dbf6d4b; sid=5yvjlnfi; browser_resolution=1920-971; bili_ticket=eyJhbGciOiJFUzM4NCIsImtpZCI6ImVjMDIiLCJ0eXAiOiJKV1QifQ.eyJleHAiOjE2OTMzNjY1MTcsImlhdCI6MTY5MzEwNzMxNywicGx0IjotMX0.I1Yfp8S9UIkU4S0G5vtBJfslPtgY7QLCj1dx9WQpyRmxKpZoA1qB5UYXNW4KBSZFGljMm7F1lbGXSGco7F79JZJ2sZNBvH9QiSVlmipzAJKaucIoFh6s3m1jpqjLp10r; bili_ticket_expires=1693366517; bp_video_offset_171300042=834376858445283367; b_lsid=1021245DB_18A3567E5C2; CURRENT_QUALITY=80; PVID=2"
    }

    # 发送请求,得到响应对象
    response_ = requests.get(url_, headers=headers_)

    str_data = response_.text  # 视频主页的html代码,类型是字符串

    # 使用xpath解析html代码,,得到想要的url
    html_obj = etree.HTML(str_data)  # 转换格式类型

    # 获取视频的名称
    res_ = html_obj.xpath('//title/text()')[0]
    # 视频名称的获取
    title_ = re.findall(r'(.*?)_哔哩哔哩', res_)[0]
    # 影响视频合成的特殊字符的处理，目前就遇到过这三个，实际上很有可能不止这三个，遇到了就用同样的方法处理就好了
    title_ = title_.replace('/', '')
    title_ = title_.replace(' ', '')
    title_ = title_.replace('&', '')
    title_ = title_.replace(':', '')

    # 使用xpath语法获取数据,取到数据为列表,索引[0]取值取出里面的字符串,即包含视频音频文件的url字符串
    url_list_str = html_obj.xpath('//script[contains(text(),"window.__playinfo__")]/text()')[0]

    # 纯视频的url
    video_url = re.findall(r'"video":\[{"id":\d+,"baseUrl":"(.*?)"', url_list_str)[0]

    # 纯音频的url
    audio_url = re.findall(r'"audio":\[{"id":\d+,"baseUrl":"(.*?)"', url_list_str)[0]

    # 设置跳转字段的headers
    headers_ = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.111 Safari/537.36',
        'Referer': url_
    }

    # 获取纯视频的数据
    response_video = requests.get(video_url, headers=headers_, stream=True)
    bytes_video = response_video.content
    # 获取纯音频的数据
    response_audio = requests.get(audio_url, headers=headers_, stream=True)
    bytes_audio = response_audio.content

    # 获取文件大小, 单位为KB
    video_size = int(int(response_video.headers['content-length']) / 1024)
    audio_size = int(int(response_audio.headers['content-length']) / 1024)

    # 保存纯视频的文件
    title_1 = title_ + '!'  # 名称进行修改,避免重名
    title_1 = title_1.replace(':', '_')
    
    with open(f'{title_1}.mp4', 'wb') as f:
        f.write(bytes_video)
        # print(f'{title_1}纯视频文件下载完毕...,大小为:{video_size}KB, {int(video_size/1024)}MB')

    with open(f'{title_1}.mp3', 'wb') as f:
        f.write(bytes_audio)
        # print(f'{title_1}纯音频文件下载完毕...,大小为:{audio_size}KB, {int(audio_size/1024)}MB')

        # 利用第三方工具ffmpeg 合成视频, 需要执行终端命令
    ffmpeg_path = r".\ffmpeg\bin\ffmpeg.exe"
    # os.system(f'{ffmpeg_path} -i {title_1}.mp3 -i {title_1}.mp4 -c copy .\video\{title_}.mp4 -loglevel quiet')


    folder_path = f"./video/{title_}"  # 替换为你想要创建的文件夹路径

    if not os.path.exists(folder_path):
        os.mkdir(folder_path)
        # print(f"The folder '{folder_path}' already exists.")


    command = f'{ffmpeg_path} -i {title_1}.mp3 -i {title_1}.mp4 -c copy ./video/{title_}/{title_}.mp4 -loglevel quiet'

    os.system(command)


    # 显示合成文件的大小

    print(f'{title_}  下载完成')

    # 移除纯视频文件,
    os.remove(f'{title_1}.mp4')
    # 移除纯音频文件,
    os.remove(f'{title_1}.mp3')

3.2 下载选集视频

选集视频的播放链接很好找，就是后面的p=几啥的，拼一下就可以拿到整个的播放链接了
代码

import requests
import os
from lxml import etree
import re

# 获取网页源码
def getUrls2(url):
    # 发送请求,得到响应对象
        # 设置用户代理,cookie
    headers = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36',
        'Cookie': "buvid3=7014DDC0-BF1E-B121-F5A5-F10753C840B423630infoc; i-wanna-go-back=-1; _uuid=49BF2138-1E10F-D5F5-10898-D8311651B53927883infoc; FEED_LIVE_VERSION=V8; DedeUserID=171300042; DedeUserID__ckMd5=c65bec3211413192; CURRENT_FNVAL=4048; rpdid=|(J|)J~m~llk0J'uYm|)~klRl; header_theme_version=CLOSE; hit-new-style-dyn=1; hit-dyn-v2=1; is-2022-channel=1; fingerprint=fe5c7462625770aa2abce449a7c01fd2; buvid_fp_plain=undefined; b_nut=1691207170; b_ut=5; buvid_fp=fe5c7462625770aa2abce449a7c01fd2; LIVE_BUVID=AUTO4016915564967297; buvid4=1AE73807-AEA0-7078-DA57-7F9FE5C3D6F896987-023080912-A0g5nInZwV3VmJJT68FJxw%3D%3D; home_feed_column=5; SESSDATA=fc1266d3%2C1708653865%2C29c08%2A81-i-T9HQrucvpCVcPwSwXl5LmjTyduIzF9veu0KS9i2IwXK_xkcqlt1XQyxJ3sG-9HMSwLwAAKgA; bili_jct=068bc0a79f3fa7aa1a030e478dbf6d4b; sid=5yvjlnfi; browser_resolution=1920-971; bili_ticket=eyJhbGciOiJFUzM4NCIsImtpZCI6ImVjMDIiLCJ0eXAiOiJKV1QifQ.eyJleHAiOjE2OTMzNjY1MTcsImlhdCI6MTY5MzEwNzMxNywicGx0IjotMX0.I1Yfp8S9UIkU4S0G5vtBJfslPtgY7QLCj1dx9WQpyRmxKpZoA1qB5UYXNW4KBSZFGljMm7F1lbGXSGco7F79JZJ2sZNBvH9QiSVlmipzAJKaucIoFh6s3m1jpqjLp10r; bili_ticket_expires=1693366517; bp_video_offset_171300042=834376858445283367; b_lsid=1021245DB_18A3567E5C2; CURRENT_QUALITY=80; PVID=2"
    }
    response_ = requests.get(url, headers=headers)

    str_data = response_.text  # 视频主页的html代码,类型是字符串

    # 使用xpath解析html代码,,得到想要的url
    html_obj = etree.HTML(str_data)  # 转换格式类型
    urls = []
    # 获取了li的数量，
    lis = html_obj.xpath("//ul[@class='list-box']/li")
    question_mark_index = url.find('?')

    # 如果找到了 '?'，就截取该位置之前的子串
    if question_mark_index != -1:
        cleaned_url = url[:question_mark_index]
    else:
        cleaned_url = url
    # print(cleaned_url)
    # 拼接api
    for i in range(1,len(lis)+1):
        # print(i)
        strs = cleaned_url + "?p=" + str(i)
        urls.append(strs)

    # print(content)
    return urls

import requests
import os
from lxml import etree
import re

def videoDownload3(url_,i,name):
    # 设置用户代理,cookie
    headers_ = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36',
        'Cookie': "buvid3=7014DDC0-BF1E-B121-F5A5-F10753C840B423630infoc; i-wanna-go-back=-1; _uuid=49BF2138-1E10F-D5F5-10898-D8311651B53927883infoc; FEED_LIVE_VERSION=V8; DedeUserID=171300042; DedeUserID__ckMd5=c65bec3211413192; CURRENT_FNVAL=4048; rpdid=|(J|)J~m~llk0J'uYm|)~klRl; header_theme_version=CLOSE; hit-new-style-dyn=1; hit-dyn-v2=1; is-2022-channel=1; fingerprint=fe5c7462625770aa2abce449a7c01fd2; buvid_fp_plain=undefined; b_nut=1691207170; b_ut=5; buvid_fp=fe5c7462625770aa2abce449a7c01fd2; LIVE_BUVID=AUTO4016915564967297; buvid4=1AE73807-AEA0-7078-DA57-7F9FE5C3D6F896987-023080912-A0g5nInZwV3VmJJT68FJxw%3D%3D; home_feed_column=5; SESSDATA=fc1266d3%2C1708653865%2C29c08%2A81-i-T9HQrucvpCVcPwSwXl5LmjTyduIzF9veu0KS9i2IwXK_xkcqlt1XQyxJ3sG-9HMSwLwAAKgA; bili_jct=068bc0a79f3fa7aa1a030e478dbf6d4b; sid=5yvjlnfi; browser_resolution=1920-971; bili_ticket=eyJhbGciOiJFUzM4NCIsImtpZCI6ImVjMDIiLCJ0eXAiOiJKV1QifQ.eyJleHAiOjE2OTMzNjY1MTcsImlhdCI6MTY5MzEwNzMxNywicGx0IjotMX0.I1Yfp8S9UIkU4S0G5vtBJfslPtgY7QLCj1dx9WQpyRmxKpZoA1qB5UYXNW4KBSZFGljMm7F1lbGXSGco7F79JZJ2sZNBvH9QiSVlmipzAJKaucIoFh6s3m1jpqjLp10r; bili_ticket_expires=1693366517; bp_video_offset_171300042=834376858445283367; b_lsid=1021245DB_18A3567E5C2; CURRENT_QUALITY=80; PVID=2"
    }

    # 发送请求,得到响应对象
    response_ = requests.get(url_, headers=headers_)

    str_data = response_.text  # 视频主页的html代码,类型是字符串

    # 使用xpath解析html代码,,得到想要的url
    html_obj = etree.HTML(str_data)  # 转换格式类型

    # 获取视频的名称
    res_ = html_obj.xpath('//title/text()')[0]
    # 视频名称的获取
    title_ = re.findall(r'(.*?)_哔哩哔哩', res_)[0]
    
    fileName = name
    # 影响视频合成的特殊字符的处理，目前就遇到过这三个，实际上很有可能不止这三个，遇到了就用同样的方法处理就好了
    title_ = title_.replace('/', '')
    title_ = title_.replace(' ', '')
    title_ = title_.replace('&', '')
    title_ = title_.replace(':', '')
    title_ = title_.replace('-', '')
    title_ = title_.replace('—', '')
    # 使用xpath语法获取数据,取到数据为列表,索引[0]取值取出里面的字符串,即包含视频音频文件的url字符串
    url_list_str = html_obj.xpath('//script[contains(text(),"window.__playinfo__")]/text()')[0]

    # 纯视频的url
    video_url = re.findall(r'"video":\[{"id":\d+,"baseUrl":"(.*?)"', url_list_str)[0]

    # 纯音频的url
    audio_url = re.findall(r'"audio":\[{"id":\d+,"baseUrl":"(.*?)"', url_list_str)[0]

    # 设置跳转字段的headers
    headers_ = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.111 Safari/537.36',
        'Referer': url_
    }

    # 获取纯视频的数据
    response_video = requests.get(video_url, headers=headers_, stream=True)
    bytes_video = response_video.content
    # 获取纯音频的数据
    response_audio = requests.get(audio_url, headers=headers_, stream=True)
    bytes_audio = response_audio.content

    # 获取文件大小, 单位为KB
    video_size = int(int(response_video.headers['content-length']) / 1024)
    audio_size = int(int(response_audio.headers['content-length']) / 1024)

    # 保存纯视频的文件
    title_1 = title_ + '!'  # 名称进行修改,避免重名
    title_1 = title_1.replace(':', '')
    with open(f'{title_1}.mp4', 'wb') as f:
        f.write(bytes_video)
        # print(f'{title_1}纯视频文件下载完毕...,大小为:{video_size}KB, {int(video_size/1024)}MB')

    with open(f'{title_1}.mp3', 'wb') as f:
        f.write(bytes_audio)
        # print(f'{title_1}纯音频文件下载完毕...,大小为:{audio_size}KB, {int(audio_size/1024)}MB')

        # 利用第三方工具ffmpeg 合成视频, 需要执行终端命令
    ffmpeg_path = r".\ffmpeg\bin\ffmpeg.exe"
    # os.system(f'{ffmpeg_path} -i {title_1}.mp3 -i {title_1}.mp4 -c copy .\video\{title_}.mp4 -loglevel quiet')

    folder_path = f"./video/{fileName}"  # 替换为你想要创建的文件夹路径
    if not os.path.exists(folder_path):
        os.mkdir(folder_path)
        # print(f"The folder '{folder_path}' already exists.")

    command = f'{ffmpeg_path} -i {title_1}.mp3 -i {title_1}.mp4 -c copy ./video/{fileName}/{i}.{title_1}.mp4 -loglevel quiet'

    file_path = f"./video/{fileName}/{i}.{title_}.mp4"
    if os.path.exists(file_path):
        pass
    else:
        os.system(command)

    # 显示合成文件的大小

    print(f'{i}.{title_}  下载完成')

    # 移除纯视频文件,
    os.remove(f'{title_1}.mp4')
    # 移除纯音频文件,
    os.remove(f'{title_1}.mp3')

3.3 下载合集视频

合集的里面数据的访问api
python爬取bilibili，下载视频,python,python,爬虫
合集里面的数据，就是从这个里面拿到播放id，给json中的处理拿出来，拼接视频播放链接

代码

# 获取网页源码
def getUrls3(url):
    # 发送请求,得到响应对象
        # 设置用户代理,cookie
    headers = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36',
        'Cookie': "buvid3=7014DDC0-BF1E-B121-F5A5-F10753C840B423630infoc; i-wanna-go-back=-1; _uuid=49BF2138-1E10F-D5F5-10898-D8311651B53927883infoc; FEED_LIVE_VERSION=V8; DedeUserID=171300042; DedeUserID__ckMd5=c65bec3211413192; CURRENT_FNVAL=4048; rpdid=|(J|)J~m~llk0J'uYm|)~klRl; header_theme_version=CLOSE; hit-new-style-dyn=1; hit-dyn-v2=1; is-2022-channel=1; fingerprint=fe5c7462625770aa2abce449a7c01fd2; buvid_fp_plain=undefined; b_nut=1691207170; b_ut=5; buvid_fp=fe5c7462625770aa2abce449a7c01fd2; LIVE_BUVID=AUTO4016915564967297; buvid4=1AE73807-AEA0-7078-DA57-7F9FE5C3D6F896987-023080912-A0g5nInZwV3VmJJT68FJxw%3D%3D; home_feed_column=5; SESSDATA=fc1266d3%2C1708653865%2C29c08%2A81-i-T9HQrucvpCVcPwSwXl5LmjTyduIzF9veu0KS9i2IwXK_xkcqlt1XQyxJ3sG-9HMSwLwAAKgA; bili_jct=068bc0a79f3fa7aa1a030e478dbf6d4b; sid=5yvjlnfi; browser_resolution=1920-971; bili_ticket=eyJhbGciOiJFUzM4NCIsImtpZCI6ImVjMDIiLCJ0eXAiOiJKV1QifQ.eyJleHAiOjE2OTMzNjY1MTcsImlhdCI6MTY5MzEwNzMxNywicGx0IjotMX0.I1Yfp8S9UIkU4S0G5vtBJfslPtgY7QLCj1dx9WQpyRmxKpZoA1qB5UYXNW4KBSZFGljMm7F1lbGXSGco7F79JZJ2sZNBvH9QiSVlmipzAJKaucIoFh6s3m1jpqjLp10r; bili_ticket_expires=1693366517; bp_video_offset_171300042=834376858445283367; b_lsid=1021245DB_18A3567E5C2; CURRENT_QUALITY=80; PVID=2"
    }
    # 使用正则表达式提取数字
    pattern = r'\d+'
    numbers = re.findall(pattern, url)
    mid = numbers[0]
    season_id = numbers[1]
    page_num = 1
    url = f"https://api.bilibili.com/x/polymer/web-space/seasons_archives_list?mid={mid}&season_id={season_id}&sort_reverse=false&page_num={page_num}&page_size=30"
    response = requests.get(url)

    if response.status_code == 200:
        json_data = response.json()
        # print(json_data["data"]["page"]["total"])
        total = int(json_data["data"]["page"]["total"])
        page_size = int(json_data["data"]["page"]["page_size"])
        page = int(total / page_size) + 1
        name = json_data["data"]["meta"]["name"]
        # print(total,page)
    
    urls = []
    # 
    for i in range(1,page+1):
        # print(i) 
        url = f"https://api.bilibili.com/x/polymer/web-space/seasons_archives_list?mid={mid}&season_id={season_id}&sort_reverse=false&page_num={i}&page_size=30"
        response = requests.get(url)
        if response.status_code == 200:
            json_data = response.json()
            archives = json_data["data"]["archives"]
            num = 0
            for j in archives:
                bvid = archives[num]["bvid"]
                videoUrl = f"https://www.bilibili.com/video/{bvid}/"
                num = num + 1
                urls.append(videoUrl)
    return urls,name

import requests
import os
from lxml import etree
import re

def videoDownload2(url_,i):
    # 设置用户代理,cookie
    headers_ = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36',
        'Cookie': "buvid3=7014DDC0-BF1E-B121-F5A5-F10753C840B423630infoc; i-wanna-go-back=-1; _uuid=49BF2138-1E10F-D5F5-10898-D8311651B53927883infoc; FEED_LIVE_VERSION=V8; DedeUserID=171300042; DedeUserID__ckMd5=c65bec3211413192; CURRENT_FNVAL=4048; rpdid=|(J|)J~m~llk0J'uYm|)~klRl; header_theme_version=CLOSE; hit-new-style-dyn=1; hit-dyn-v2=1; is-2022-channel=1; fingerprint=fe5c7462625770aa2abce449a7c01fd2; buvid_fp_plain=undefined; b_nut=1691207170; b_ut=5; buvid_fp=fe5c7462625770aa2abce449a7c01fd2; LIVE_BUVID=AUTO4016915564967297; buvid4=1AE73807-AEA0-7078-DA57-7F9FE5C3D6F896987-023080912-A0g5nInZwV3VmJJT68FJxw%3D%3D; home_feed_column=5; SESSDATA=fc1266d3%2C1708653865%2C29c08%2A81-i-T9HQrucvpCVcPwSwXl5LmjTyduIzF9veu0KS9i2IwXK_xkcqlt1XQyxJ3sG-9HMSwLwAAKgA; bili_jct=068bc0a79f3fa7aa1a030e478dbf6d4b; sid=5yvjlnfi; browser_resolution=1920-971; bili_ticket=eyJhbGciOiJFUzM4NCIsImtpZCI6ImVjMDIiLCJ0eXAiOiJKV1QifQ.eyJleHAiOjE2OTMzNjY1MTcsImlhdCI6MTY5MzEwNzMxNywicGx0IjotMX0.I1Yfp8S9UIkU4S0G5vtBJfslPtgY7QLCj1dx9WQpyRmxKpZoA1qB5UYXNW4KBSZFGljMm7F1lbGXSGco7F79JZJ2sZNBvH9QiSVlmipzAJKaucIoFh6s3m1jpqjLp10r; bili_ticket_expires=1693366517; bp_video_offset_171300042=834376858445283367; b_lsid=1021245DB_18A3567E5C2; CURRENT_QUALITY=80; PVID=2"
    }

    # 发送请求,得到响应对象
    response_ = requests.get(url_, headers=headers_)

    str_data = response_.text  # 视频主页的html代码,类型是字符串

    # 使用xpath解析html代码,,得到想要的url
    html_obj = etree.HTML(str_data)  # 转换格式类型

    # 获取视频的名称
    res_ = html_obj.xpath('//title/text()')[0]
    # 视频名称的获取
    title_ = re.findall(r'(.*?)_哔哩哔哩', res_)[0]
    
    fileName = html_obj.xpath('//h1[@class="video-title"]/text()')[0]
    
    # 影响视频合成的特殊字符的处理，目前就遇到过这三个，实际上很有可能不止这三个，遇到了就用同样的方法处理就好了
    title_ = title_.replace('/', '')
    title_ = title_.replace(' ', '')
    title_ = title_.replace('&', '')
    title_ = title_.replace(':', '')

    # 使用xpath语法获取数据,取到数据为列表,索引[0]取值取出里面的字符串,即包含视频音频文件的url字符串
    url_list_str = html_obj.xpath('//script[contains(text(),"window.__playinfo__")]/text()')[0]

    # 纯视频的url
    video_url = re.findall(r'"video":\[{"id":\d+,"baseUrl":"(.*?)"', url_list_str)[0]

    # 纯音频的url
    audio_url = re.findall(r'"audio":\[{"id":\d+,"baseUrl":"(.*?)"', url_list_str)[0]

    # 设置跳转字段的headers
    headers_ = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.111 Safari/537.36',
        'Referer': url_
    }

    # 获取纯视频的数据
    response_video = requests.get(video_url, headers=headers_, stream=True)
    bytes_video = response_video.content
    # 获取纯音频的数据
    response_audio = requests.get(audio_url, headers=headers_, stream=True)
    bytes_audio = response_audio.content

    # 获取文件大小, 单位为KB
    video_size = int(int(response_video.headers['content-length']) / 1024)
    audio_size = int(int(response_audio.headers['content-length']) / 1024)

    # 保存纯视频的文件
    title_1 = title_ + '!'  # 名称进行修改,避免重名
    title_1 = title_1.replace(':', '_')
    with open(f'{title_1}.mp4', 'wb') as f:
        f.write(bytes_video)
        # print(f'{title_1}纯视频文件下载完毕...,大小为:{video_size}KB, {int(video_size/1024)}MB')

    with open(f'{title_1}.mp3', 'wb') as f:
        f.write(bytes_audio)
        # print(f'{title_1}纯音频文件下载完毕...,大小为:{audio_size}KB, {int(audio_size/1024)}MB')

        # 利用第三方工具ffmpeg 合成视频, 需要执行终端命令
    ffmpeg_path = r".\ffmpeg\bin\ffmpeg.exe"
    # os.system(f'{ffmpeg_path} -i {title_1}.mp3 -i {title_1}.mp4 -c copy .\video\{title_}.mp4 -loglevel quiet')

    folder_path = f"./video/{fileName}"  # 替换为你想要创建的文件夹路径
    if not os.path.exists(folder_path):
        os.mkdir(folder_path)
        # print(f"The folder '{folder_path}' already exists.")

    command = f'{ffmpeg_path} -i {title_1}.mp3 -i {title_1}.mp4 -c copy ./video/{fileName}/{i}.{title_}.mp4 -loglevel quiet'

    file_path = f"./video/{fileName}/{i}.{title_}.mp4"
    if os.path.exists(file_path):
        pass
    else:
        os.system(command)

    # 显示合成文件的大小

    print(f'{i}.{title_}  下载完成')

    # 移除纯视频文件,
    os.remove(f'{title_1}.mp4')
    # 移除纯音频文件,
    os.remove(f'{title_1}.mp3')

3.4 多线程

代码

import concurrent.futures
import requests

# 定义一个下载函数
def download_video(URL):
    url, index, name = URL.split(" ", 2)
    videoDownload3(url,index,name)

def THREAD(URLS):
    # 创建线程池，指定线程数量
    max_workers = 10  # 这里设置线程数量，根据需要进行调整
    with concurrent.futures.ThreadPoolExecutor(max_workers=max_workers) as executor:
        # 提交下载任务给线程池处理
        futures = [executor.submit(download_video, URL) for URL in URLS]

        # 等待所有任务完成
        for future in concurrent.futures.as_completed(futures):
            try:
                future.result()  # 获取任务的结果（这里不需要结果）
            except Exception as e:
                print(f"An error occurred: {e}")

3.5 结果

url_model = "https://space.bilibili.com/471303350/channel/collectiondetail?sid=1278346 3"
value = url_model.split(' ')
url = value[0]
model = value[1]

if model == "1":
    videoDownload1(url)
    print("下载完成")
if model == "2":
    # 接口分析
    # 点进去的话接口
    # https://www.bilibili.com/video/BV1qW4y1a7fU/?spm_id_from=333.337.search-card.all.click
    # 点击视频的话就这样
    # https://www.bilibili.com/video/BV1qW4y1a7fU?p=1
    # https://www.bilibili.com/video/BV1qW4y1a7fU?p=2&vd_source=de2dcd0f37ff916ec3f8fb83c6366123
    # 可以发现不同的集的接口格式应该是这样的，p = 几就是第几集
    # https://www.bilibili.com/video/BV1qW4y1a7fU?p=1

    # 查看有多少集
    # 一种是视频选集那块会写有多少个
    # 获取源码
    urls = getUrls2(url)
    i = 1
    for index,url in enumerate(urls):
        videoDownload2(url,index)

    print("下载完成")
if model == "3":
    # 接口分析
    # 视频合计每个视频接口没有规律，然后再播放页中网页没有直接的播放链接，所以就用合集页的链接来分析
    # 网页里面的每个链接都是动态加载的，需要访问json数据获取，也或者用虚拟浏览器那种等页面加载完成后访问(这种以后可能会更新，感觉这个有点麻烦)，
    # 这里是用json数据做的
    # https://space.bilibili.com/107762251/channel/collectiondetail?sid=877119
    # https://api.bilibili.com/x/polymer/web-space/seasons_archives_list?mid=107762251&season_id=877119&sort_reverse=false&page_num=1&page_size=30
    # https://space.bilibili.com/389199842/channel/collectiondetail?sid=1275285
    # https://api.bilibili.com/x/polymer/web-space/seasons_archives_list?mid=389199842&season_id=1275285&sort_reverse=false&page_num=1&page_size=30
    # 这是两个接口，前面那个数字是用户，后面那个数字代表的是合集，下载的接口其实是股东

    urls,name = getUrls3(url)
    # print(len(urls))
    for index,url in enumerate(urls):
        # print(url)
        videoDownload3(url,index,name)
    # print(urls)
    # 多线程
    # for index,url in enumerate(urls):
    #     URLS.append(url + " " + str(index) + " " + name)
    #     THREAD(URLS)

那切里做展示，有些合集下载时候有点bug，还没找到问题，可以下载，但是保存路径有点问题，应该是和命令行冲突了，我就不改了
python爬取bilibili，下载视频,python,python,爬虫

3.6 合集视频更新

原来会出现部分合集显示下载成功，但是文件夹里面没有东西，是因为有些合集名字在命令里面没办法执行，因为一些特殊符号什么的，所以把合集名字手动指定一下下载就可以了，然后多线程加上去，代码如下
拿视频链接的

# 获取网页源码
def getUrls3(url):
    # 发送请求,得到响应对象
        # 设置用户代理,cookie
    headers = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36',
        'Cookie': "buvid3=7014DDC0-BF1E-B121-F5A5-F10753C840B423630infoc; i-wanna-go-back=-1; _uuid=49BF2138-1E10F-D5F5-10898-D8311651B53927883infoc; FEED_LIVE_VERSION=V8; DedeUserID=171300042; DedeUserID__ckMd5=c65bec3211413192; CURRENT_FNVAL=4048; rpdid=|(J|)J~m~llk0J'uYm|)~klRl; header_theme_version=CLOSE; hit-new-style-dyn=1; hit-dyn-v2=1; is-2022-channel=1; fingerprint=fe5c7462625770aa2abce449a7c01fd2; buvid_fp_plain=undefined; b_nut=1691207170; b_ut=5; buvid_fp=fe5c7462625770aa2abce449a7c01fd2; LIVE_BUVID=AUTO4016915564967297; buvid4=1AE73807-AEA0-7078-DA57-7F9FE5C3D6F896987-023080912-A0g5nInZwV3VmJJT68FJxw%3D%3D; home_feed_column=5; SESSDATA=fc1266d3%2C1708653865%2C29c08%2A81-i-T9HQrucvpCVcPwSwXl5LmjTyduIzF9veu0KS9i2IwXK_xkcqlt1XQyxJ3sG-9HMSwLwAAKgA; bili_jct=068bc0a79f3fa7aa1a030e478dbf6d4b; sid=5yvjlnfi; browser_resolution=1920-971; bili_ticket=eyJhbGciOiJFUzM4NCIsImtpZCI6ImVjMDIiLCJ0eXAiOiJKV1QifQ.eyJleHAiOjE2OTMzNjY1MTcsImlhdCI6MTY5MzEwNzMxNywicGx0IjotMX0.I1Yfp8S9UIkU4S0G5vtBJfslPtgY7QLCj1dx9WQpyRmxKpZoA1qB5UYXNW4KBSZFGljMm7F1lbGXSGco7F79JZJ2sZNBvH9QiSVlmipzAJKaucIoFh6s3m1jpqjLp10r; bili_ticket_expires=1693366517; bp_video_offset_171300042=834376858445283367; b_lsid=1021245DB_18A3567E5C2; CURRENT_QUALITY=80; PVID=2"
    }
    # 使用正则表达式提取数字
    pattern = r'\d+'
    numbers = re.findall(pattern, url)
    mid = numbers[0]
    season_id = numbers[1]
    page_num = 1
    url = f"https://api.bilibili.com/x/polymer/web-space/seasons_archives_list?mid={mid}&season_id={season_id}&sort_reverse=false&page_num={page_num}&page_size=30"
    response = requests.get(url)

    if response.status_code == 200:
        json_data = response.json()
        # print(json_data["data"]["page"]["total"])
        total = int(json_data["data"]["page"]["total"])
        page_size = int(json_data["data"]["page"]["page_size"])
        page = int(total / page_size) + 1
        name = json_data["data"]["meta"]["name"]
        # print(total,page)
    
    urls = []
    # 
    for i in range(1,page+1):
        # print(i) 
        url = f"https://api.bilibili.com/x/polymer/web-space/seasons_archives_list?mid={mid}&season_id={season_id}&sort_reverse=false&page_num={i}&page_size=30"
        response = requests.get(url)
        if response.status_code == 200:
            json_data = response.json()
            archives = json_data["data"]["archives"]
            num = 0
            for j in archives:
                bvid = archives[num]["bvid"]
                videoUrl = f"https://www.bilibili.com/video/{bvid}/"
                num = num + 1
                urls.append(videoUrl)
    return urls,name

下载视频的

import requests
import os
from lxml import etree
import re

def videoDownload3(url_,index,name):
    # 设置用户代理,cookie
    headers_ = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36',
        'Cookie': "buvid3=7014DDC0-BF1E-B121-F5A5-F10753C840B423630infoc; i-wanna-go-back=-1; _uuid=49BF2138-1E10F-D5F5-10898-D8311651B53927883infoc; FEED_LIVE_VERSION=V8; DedeUserID=171300042; DedeUserID__ckMd5=c65bec3211413192; CURRENT_FNVAL=4048; rpdid=|(J|)J~m~llk0J'uYm|)~klRl; header_theme_version=CLOSE; hit-new-style-dyn=1; hit-dyn-v2=1; is-2022-channel=1; fingerprint=fe5c7462625770aa2abce449a7c01fd2; buvid_fp_plain=undefined; b_nut=1691207170; b_ut=5; buvid_fp=fe5c7462625770aa2abce449a7c01fd2; LIVE_BUVID=AUTO4016915564967297; buvid4=1AE73807-AEA0-7078-DA57-7F9FE5C3D6F896987-023080912-A0g5nInZwV3VmJJT68FJxw%3D%3D; home_feed_column=5; SESSDATA=fc1266d3%2C1708653865%2C29c08%2A81-i-T9HQrucvpCVcPwSwXl5LmjTyduIzF9veu0KS9i2IwXK_xkcqlt1XQyxJ3sG-9HMSwLwAAKgA; bili_jct=068bc0a79f3fa7aa1a030e478dbf6d4b; sid=5yvjlnfi; browser_resolution=1920-971; bili_ticket=eyJhbGciOiJFUzM4NCIsImtpZCI6ImVjMDIiLCJ0eXAiOiJKV1QifQ.eyJleHAiOjE2OTMzNjY1MTcsImlhdCI6MTY5MzEwNzMxNywicGx0IjotMX0.I1Yfp8S9UIkU4S0G5vtBJfslPtgY7QLCj1dx9WQpyRmxKpZoA1qB5UYXNW4KBSZFGljMm7F1lbGXSGco7F79JZJ2sZNBvH9QiSVlmipzAJKaucIoFh6s3m1jpqjLp10r; bili_ticket_expires=1693366517; bp_video_offset_171300042=834376858445283367; b_lsid=1021245DB_18A3567E5C2; CURRENT_QUALITY=80; PVID=2"
    }

    # 发送请求,得到响应对象
    response_ = requests.get(url_, headers=headers_)

    str_data = response_.text  # 视频主页的html代码,类型是字符串

    # 使用xpath解析html代码,,得到想要的url
    html_obj = etree.HTML(str_data)  # 转换格式类型

    # 获取视频的名称
    res_ = html_obj.xpath('//title/text()')[0]
    # 视频名称的获取
    title_ = re.findall(r'(.*?)_哔哩哔哩', res_)[0]
    # 影响视频合成的特殊字符的处理，目前就遇到过这三个，实际上很有可能不止这三个，遇到了就用同样的方法处理就好了
    title_ = title_.replace('/', '')
    title_ = title_.replace(' ', '')
    title_ = title_.replace('&', '')
    title_ = title_.replace(':', '')

    # 使用xpath语法获取数据,取到数据为列表,索引[0]取值取出里面的字符串,即包含视频音频文件的url字符串
    url_list_str = html_obj.xpath('//script[contains(text(),"window.__playinfo__")]/text()')[0]

    # 纯视频的url
    video_url = re.findall(r'"video":\[{"id":\d+,"baseUrl":"(.*?)"', url_list_str)[0]

    # 纯音频的url
    audio_url = re.findall(r'"audio":\[{"id":\d+,"baseUrl":"(.*?)"', url_list_str)[0]

    # 设置跳转字段的headers
    headers_ = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.111 Safari/537.36',
        'Referer': url_
    }

    # 获取纯视频的数据
    response_video = requests.get(video_url, headers=headers_, stream=True)
    bytes_video = response_video.content
    # 获取纯音频的数据
    response_audio = requests.get(audio_url, headers=headers_, stream=True)
    bytes_audio = response_audio.content

    # 获取文件大小, 单位为KB
    video_size = int(int(response_video.headers['content-length']) / 1024)
    audio_size = int(int(response_audio.headers['content-length']) / 1024)

    # 保存纯视频的文件
    title_1 = title_ + '!'  # 名称进行修改,避免重名
    title_1 = title_1.replace(':', '_')
    
    with open(f'{title_1}.mp4', 'wb') as f:
        f.write(bytes_video)
        # print(f'{title_1}纯视频文件下载完毕...,大小为:{video_size}KB, {int(video_size/1024)}MB')

    with open(f'{title_1}.mp3', 'wb') as f:
        f.write(bytes_audio)
        # print(f'{title_1}纯音频文件下载完毕...,大小为:{audio_size}KB, {int(audio_size/1024)}MB')

        # 利用第三方工具ffmpeg 合成视频, 需要执行终端命令
    ffmpeg_path = r".\ffmpeg\bin\ffmpeg.exe"
    # os.system(f'{ffmpeg_path} -i {title_1}.mp3 -i {title_1}.mp4 -c copy .\video\{title_}.mp4 -loglevel quiet')


    folder_path = f"./video/{name}"  # 替换为你想要创建的文件夹路径

    if not os.path.exists(folder_path):
        os.mkdir(folder_path)
        # print(f"The folder '{folder_path}' already exists.")


    command = f'{ffmpeg_path} -i {title_1}.mp3 -i {title_1}.mp4 -c copy ./video/{name}/{index}.{title_}.mp4 -loglevel quiet'

    os.system(command)


    # 显示合成文件的大小

    print(f'{title_}  下载完成')

    # 移除纯视频文件,
    os.remove(f'{title_1}.mp4')
    # 移除纯音频文件,
    os.remove(f'{title_1}.mp3')

多线程

import concurrent.futures
import requests

# 定义一个下载函数
def download_video(URL):
    url, index, name = URL.split(" ", 2)
    videoDownload3(url,index,name)

def THREAD(URLS):
    # 创建线程池，指定线程数量
    max_workers = 10  # 这里设置线程数量，根据需要进行调整
    with concurrent.futures.ThreadPoolExecutor(max_workers=max_workers) as executor:
        # 提交下载任务给线程池处理
        futures = [executor.submit(download_video, URL) for URL in URLS]

        # 等待所有任务完成
        for future in concurrent.futures.as_completed(futures):
            try:
                future.result()  # 获取任务的结果（这里不需要结果）
            except Exception as e:
                print(f"An error occurred: {e}")

执行


url_model = "https://space.bilibili.com/389199842/channel/collectiondetail?sid=1275285 3"
value = url_model.split(' ')
url = value[0]
model = value[1]

if model == "1":
    videoDownload1(url)
    print("下载完成")
if model == "2":
    # 接口分析
    # 点进去的话接口
    # https://www.bilibili.com/video/BV1qW4y1a7fU/?spm_id_from=333.337.search-card.all.click
    # 点击视频的话就这样
    # https://www.bilibili.com/video/BV1qW4y1a7fU?p=1
    # https://www.bilibili.com/video/BV1qW4y1a7fU?p=2&vd_source=de2dcd0f37ff916ec3f8fb83c6366123
    # 可以发现不同的集的接口格式应该是这样的，p = 几就是第几集
    # https://www.bilibili.com/video/BV1qW4y1a7fU?p=1

    # 查看有多少集
    # 一种是视频选集那块会写有多少个
    # 获取源码
    urls = getUrls2(url)
    i = 1
    for index,url in enumerate(urls):
        videoDownload2(url,index)

    print("下载完成")
if model == "3":
    # 接口分析
    # 视频合计每个视频接口没有规律，然后再播放页中网页没有直接的播放链接，所以就用合集页的链接来分析
    # 网页里面的每个链接都是动态加载的，需要访问json数据获取，也或者用虚拟浏览器那种等页面加载完成后访问(这种以后可能会更新，感觉这个有点麻烦)，
    # 这里是用json数据做的
    # https://space.bilibili.com/107762251/channel/collectiondetail?sid=877119
    # https://api.bilibili.com/x/polymer/web-space/seasons_archives_list?mid=107762251&season_id=877119&sort_reverse=false&page_num=1&page_size=30
    # https://space.bilibili.com/389199842/channel/collectiondetail?sid=1275285
    # https://api.bilibili.com/x/polymer/web-space/seasons_archives_list?mid=389199842&season_id=1275285&sort_reverse=false&page_num=1&page_size=30
    # 这是两个接口，前面那个数字是用户，后面那个数字代表的是合集，下载的接口其实是股东

    urls,name = getUrls3(url)
    name = "qml项目"
    URLS = []
    # print(len(urls))
    for index,url in enumerate(urls):
        # print(url)
        URLS.append(url + " " + str(index+1) + " " + name)
    THREAD(URLS)
    print("全部下载完成！！！")
    # print(urls)
    # for index,url in enumerate(urls):
    #     URLS.append(url + " " + str(index) + " " + name)
    #     THREAD(URLS)