某腾X视频下载器2.1

最新推荐文章于 2025-06-10 23:00:08 发布

摆渡搜不到你

最新推荐文章于 2025-06-10 23:00:08 发布

阅读量1.2k

点赞数 43

文章标签：音视频 python

本文链接：https://blog.csdn.net/2501_91968881/article/details/148513301

版权

import re
import os
import time
from bs4 import BeautifulSoup
from DrissionPage import ChromiumPage, ChromiumOptions
import subprocess
 
# 编码和清晰度映射表
codec_mapping = {
    'f321004': ('H264', '1080'),
    'f321003': ('H264', '720'),
    'f321002': ('H264', '480'),
    'f322016': ('HEVC', '4K'),
    'f322064': ('HEVC', '1080'),
    'f322063': ('HEVC', '720'),
    'f322062': ('HEVC', '480')
}
 
# 获取脚本所在目录
script_dir = os.path.dirname(os.path.abspath(__file__))
download_dir = os.path.join(script_dir, 'txdownload')
if not os.path.exists(download_dir):
    os.makedirs(download_dir)
 
target_url = input("请输入当前腾讯视频播放页地址（例如：https://v.qq.com/x/cover/mzcxxxxxxx/xxxxx.html）: ").strip()
 
co = ChromiumOptions()
co.headless(False)
page = ChromiumPage(addr_or_opts=co)
page.listen.start()
 
print(f"[+] 正在访问: {target_url}")
page.get(target_url)
 
downloaded_titles = set()
is_downloaded = False
last_base_url = None
current_episode = None
waiting_for_first_ts = False  # 等待第一个TS文件的标志
get_all_blacks_detected = False  # 新增：是否已检测到GetAllBlacks请求
 
def get_page_title():
    html = page.html
    soup = BeautifulSoup(html, 'html.parser')
    title_tag = soup.find('title')
    if title_tag:
        raw_title = title_tag.text.strip()
        parts = raw_title.split('_', 2)
        return f"{parts[0]}_{parts[1]}" if len(parts) >= 2 else parts[0]
    return "未命名视频"
 
def get_current_episode():
    html = page.html
    soup = BeautifulSoup(html, 'html.parser')
    # 更全面的集数选择器，适应腾讯视频的不同页面布局
    episode_selectors = [
        'div.episode-list a.current',  # 常见选择器1
        'div.mod_episode li.current',  # 常见选择器2
        'div.episode-item.active',     # 常见选择器3
        'a.episode-item.current'       # 常见选择器4
    ]
     
    for selector in episode_selectors:
        episode_element = soup.select_one(selector)
        if episode_element:
            return episode_element.get('title', '').strip() or episode_element.text.strip()
    return None
 
def process_ts_url(ts_url):
    match = re.search(r'/(\d+_)?(gzc[^/]+\.f\d{6})\.', ts_url)
    if not match:
        return None
 
    filename_part = match.group(2)
    code = filename_part.split('.')[-1]
    if code not in codec_mapping:
        return None
 
    codec, resolution = codec_mapping[code]
    protocol = ts_url.split('://')[0]
    domain_part = ts_url.split('//')[1].split('/')[0]
    path_parts = ts_url.split('//')[1].split('/')[1:-1]
    m3u8_url = f"{protocol}://{domain_part}/{'/'.join(path_parts)}/{filename_part}.ts.m3u8?ver=4"
    return m3u8_url, codec, resolution
 
def get_base_url():
    url = page.url
    return url.rsplit('.', 1)[0] if url.endswith('.html') else url
 
try:
    print("已进入自动监听模式，将持续捕获符合条件的ts文件...")
    print("请在浏览器中手动点击播放按钮，开始视频加载...")
 
    while True:
        current_base_url = get_base_url()
        new_episode = get_current_episode()
 
        if last_base_url is None:
            last_base_url = current_base_url
        if current_episode is None:
            current_episode = new_episode
 
        # 检查是否切换了集数
        if new_episode and new_episode != current_episode:
            print(f"[!] 检测到集数已切换")
            current_episode = new_episode
            is_downloaded = False
            last_base_url = current_base_url
            waiting_for_first_ts = True  # 开始等待第一个TS文件
            get_all_blacks_detected = False  # 重置GetAllBlacks检测标志
            print("[+] 等待新集数的GetAllBlacks请求和第一个TS文件...")
            continue
 
        if is_downloaded:
            if current_base_url != last_base_url:
                print("[!] 检测到播放页已更换，等待5秒缓冲加载...")
                time.sleep(5)
                last_base_url = current_base_url
                current_episode = get_current_episode()
                is_downloaded = False
                waiting_for_first_ts = True
                get_all_blacks_detected = False
            else:
                print("[!] 当前视频已下载，请切换新播放页或退出程序。")
                time.sleep(3)
                continue
 
        # 开始监听网络请求
        for packet in page.listen.steps():
            # 如果正在等待第一个TS文件
            if waiting_for_first_ts:
                # 先检测GetAllBlacks请求
                if 'GetAllBlacks' in packet.url and not get_all_blacks_detected:
                    print("[+] 已检测到集数切换后的GetAllBlacks请求，等待第一个TS文件...")
                    get_all_blacks_detected = True
                    continue
                 
                # 只有已经检测到GetAllBlacks请求后，才处理第一个TS文件
                if get_all_blacks_detected and '.ts' in packet.url and 'gzc' in packet.url:
                    print("[+] 检测到新集数的第一个TS文件请求")
                    waiting_for_first_ts = False
                    get_all_blacks_detected = False
                    # 继续处理这个TS文件
                else:
                    continue
            else:
                # 正常模式下只处理TS文件
                if not ('.ts' in packet.url and 'gzc' in packet.url):
                    continue
 
            result = process_ts_url(packet.url)
            if not result:
                continue
 
            m3u8_url, codec, resolution = result
            current_title = get_page_title()
            if current_episode:
                current_title = f"{current_title}_{current_episode}"
             
            print(f"[TS 文件请求]: {packet.url}")
            print(f"[M3U8 链接生成]: {m3u8_url}")
            print(f"  ├─ 编码格式: {codec}")
            print(f"  └─ 清晰度: {resolution}")
 
            if current_title in downloaded_titles:
                print(f"[!] 标题 '{current_title}' 已下载过，停止监听...")
                is_downloaded = True
                break
 
            choice = input(f"[?] 是否下载 '{current_title}'？(y/n): ").strip().lower()
            if choice == 'y':
                command = [
                    'N_m3u8DL-RE.exe',
                    m3u8_url,
                    '--save-name', current_title,
                    '--tmp-dir', download_dir,
                    '--save-dir', download_dir
                ]
                print("[+] 正在调用N_m3u8DL-RE下载器...")
                try:
                    subprocess.run(command, check=True)
                    print("[+] 视频下载完成！")
                    downloaded_titles.add(current_title)
                    is_downloaded = True
                except subprocess.CalledProcessError as e:
                    print(f"[-] 下载失败: {e}")
            elif choice == 'n':
                print("[!] 已跳过该视频。")
                is_downloaded = True
            else:
                print("[!] 输入无效，跳过本次操作。")
                is_downloaded = True
 
        time.sleep(1)
 
except KeyboardInterrupt:
    print("\n用户中断监听。")
finally:
    page.quit()    print("[+] 浏览器已关闭。")

成品示意图

基于ts改造m3u8原理进行监听ts文件的请求地址并修改成M3U8地址，后调用m3u8DL-RE下载器进行下载：原理如下
TS分片请求

https://defaultts.tc.qq.com/vipts.tc.qq.com/A1V3juazeTbzm0BnbB3yuBWzRPq6N_Xea19LacsVUXrQ/B_-zZfQ0GV5vY5IDrVDPNMQV3Ogj2422QtEwqDuljZ-V8m1EQ8Fz6MOJflt66DEl3utoWhXuni2lEa0UPqYzEjXOui4HKqLNk41Mbfszl3vJrd8fzaK9biyWyembtmTMP4exoA9ovWzgnKTvwo-rkq0w/svp_50112/b0xGpQp9lzQSM81mIIiD8KSR9bcJjHUbvooSEzE50F2TKOdBhywHN4u6mLtZBKBBplMhwETT0Ev0s7toJTiZ4bkBfbYmmuPNdcsP1lwdkjZR0xAVwu_yyUEYkNFDvlQiJ5Nf44-O6GfGkgHxeG_1CEkV5jUm9jpiyOaJWpjkDzR7wELeoHp29zNIKLTdhGd_UZbs_nkk_sXsQnKeXYyPMEd4bRoTPWnwszKo33CKl3c/019_gzc_1000102_0b5324breaadjaaiw2h2tzt4hv6dcliqgfsa.f322016.1.ts.index=19&start=215160&end=227080&brs=128842604&bre=135090595&ver=4&token=7e0ffa514c48c7422108f392092f781b
改：
https://defaultts.tc.qq.com/vipts.tc.qq.com/A1V3juazeTbzm0BnbB3yuBWzRPq6N_Xea19LacsVUXrQ/B_-zZfQ0GV5vY5IDrVDPNMQV3Ogj2422QtEwqDuljZ-V8m1EQ8Fz6MOJflt66DEl3utoWhXuni2lEa0UPqYzEjXOui4HKqLNk41Mbfszl3vJrd8fzaK9biyWyembtmTMP4exoA9ovWzgnKTvwo-rkq0w/svp_50112/b0xGpQp9lzQSM81mIIiD8KSR9bcJjHUbvooSEzE50F2TKOdBhywHN4u6mLtZBKBBplMhwETT0Ev0s7toJTiZ4bkBfbYmmuPNdcsP1lwdkjZR0xAVwu_yyUEYkNFDvlQiJ5Nf44-O6GfGkgHxeG_1CEkV5jUm9jpiyOaJWpjkDzR7wELeoHp29zNIKLTdhGd_UZbs_nkk_sXsQnKeXYyPMEd4bRoTPWnwszKo33CKl3c/gzc_1000102_0b5324breaadjaaiw2h2tzt4hv6dcliqgfsa.f322016.ts.m3u8?ver=4
红色部分为要修改的，亲测有效。0x0_gzc这种开头的TS分片

运行环境：Win10、Win11（win7用户可自行搞python环境自己打包，把N_m3u8DL-RE.exe和ffmpeg.exe置于脚本相同目录下）
该软件基于谷歌进行监听
需先安装谷歌浏览器！！！（尽量最新，测试环境谷歌正式版136.0.7103.114）
注：超前，付费，加密视频均不支持！！！！
已修复2.0切换集数下载重复问题，内附下载教程

下载前奏：
第一次先在浏览器上登录账号，选择最高清晰度，关闭浏览器和程序
第二次再重新打开，粘贴某个集数的视频链接
可以对该页面的所有相关视频进行监听下载！
视频自动保存在txdownload文件夹里
tx_2.1.exe为本体环境

下载链接：夸克网盘分享