B站windows客户端视频解码

B站windows客户端视频解码

import os
import json

# 只获取文件夹的列表，已排除单个文件
def get_folder_list(rootdir):
    return [os.path.join(rootdir, name) for name in os.listdir(rootdir) if os.path.isdir(os.path.join(rootdir, name))]

def fix_m4s(target_path: str, output_path: str, bufsize: int = 256 * 1024 * 1024) -> None:
    assert bufsize > 0
    with open(target_path, 'rb') as target_file:
        header = target_file.read(32)
        new_header = header.replace(b'000000000', b'')
        new_header = new_header.replace(b'$', b' ')
        new_header = new_header.replace(b'avc1', b'')
    with open(output_path, 'wb') as output_file:
        output_file.write(new_header)
        i = target_file.read(bufsize)
        while i:
            output_file.write(i)
            i = target_file.read(bufsize)

folder_list = get_folder_list("D:/Video/bilibili/")
exist_transed_folder_list = get_folder_list("D:/Video/Trans/")

# 获取所有已转换的文件的文件名
exist_transed_video_list = [
    os.path.splitext(f)[0] 
    for folder in exist_transed_folder_list 
    for f in os.listdir(folder) 
    if f.endswith('.mp3') or f.endswith('.mp4')
]

count = 0  # 计数用
total = len(folder_list) - len(exist_transed_folder_list)

for folder in folder_list:
    print(folder)  # 当前操作的目录
    
    # 找到所有的m4s文件
    m4s_files = [f"{folder}/{file}" for file in os.listdir(folder) if file.endswith('.m4s')]
    
    # 读取videoinfo信息
    with open(f"{folder}/.videoInfo", 'r', encoding='utf-8') as f:
        videoinfo = json.loads(f.read())
    bvid = videoinfo["bvid"]
    p = videoinfo["p"]
    tab_name = videoinfo["tabName"]
    
    output_dir = f"D:/Video/Trans/{bvid}"
    if not os.path.exists(output_dir):
        os.makedirs(output_dir)
    elif output_dir in exist_transed_folder_list:
        continue
    
    fix_m4s(m4s_files[0], f"{output_dir}/{str(p)}_{tab_name}.mp4")
    fix_m4s(m4s_files[1], f"{output_dir}/{str(p)}_{tab_name}.mp3")

在最后两行没对文件做判断，但是我用potplayer播放了没问题。

下载的文件里面有两个.m4s文件，文件名倒数第3位是2表示音频文件，0表示视频文件，可以自己做个判断再修改后缀。

核心代码是下面这三行替换，来源于B站的一个视频以及评论区，视频已经找不到是哪个了。

思想就是把.m4s以二进制格式打开，前9个0删掉，$改成空格，avc1删掉，就完事儿了。

new_header = header.replace(b'000000000', b'')        new_header = new_header.replace(b'$', b' ')        new_header = new_header.replace(b'avc1', b'')

放个图片做封面