使用 Python 刮削 JAV 并写入元数据

浏览本文有很可能会触犯所在地区法律，请三思而后行。

身为一名老司机，管理自己的影片库存是一件很大的事。恰好 Windows 有丰富的文件元数据管理，可将影片的演员和标题写入元数据，便能实现在文件管理器中按其排序或筛选等。

最终效果如图，需要勾选对应的元数据

我选择的的刮削网站是，因为不带封面图，加载起来比较快。需要注意，这个网站只包括 DMM 和 FANZA 的 JAV 影片，不含无码和 FC2 等。运行前需要安装 ExifTool（终端执行 winget install exiftool）和 ffmpeg（终端执行 winget install ffmpeg），并把 Windows 的语言编码改为 UTF-8（可参阅修改 windows 终端 cmd 控制台默认编码为 utf-8 - luckyangg - 博客园）。下面是代码和讲解。

import re
import os
import pathlib
import requests
import shutil
import subprocess
from bs4 import BeautifulSoup
from tkinter import filedialog

# 刮削
def scrape(movie_id):
    url = f"https://javtxt.com/search?type=id&q={movie_id}"
    headers = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64)"}

    # 设置 HTTP 或 SOCKS5 代理（根据需要修改）
    proxies = {
        "http": "http://localhost:7890",
        "https": "http://localhost:7890",
    }

    response = requests.get(url, headers=headers, proxies=proxies, timeout=10)
    if response.status_code != 200:
        print(f"请求失败，状态码: {response.status_code}")
        return None

    soup = BeautifulSoup(response.text, "html.parser")

    work = soup.find("a", class_="work")
    if not work:
        print("未找到影片信息")
        return {}

    # 标题
    title = work.find("h4", class_="work-title").text.strip()
    # 演员（多个则按,分割）
    artist = work.find("div", class_="work-actresses").text.strip()
    artist = ",".join(artist.split())

    return {"标题": title, "演员": artist}

# 正则表达式匹配番号
def recognize(s):
    pattern = r"([a-zA-Z]{2,}[-_ ]?[0-9]{2,}[-_0-9]{0,})"
    match = re.findall(pattern, s)
    if not match:
        pattern = r"([n][0-9]{4,})"
        match = re.findall(pattern, s)
        if not match:
            pattern = r"([0-9]{3,}[-_]?[0-9]{0,})"
            match = re.findall(pattern, s)

    if match:
        temp = match[-1]
        if "_" in temp:
            temp = temp.replace("_", "-")
        if temp[-1] == "-":
            temp = temp[:-1]

        # 把 abc123 修改为 abc-123
        pattern = r"^([a-zA-Z]{3,})([0-9]{3,})$"
        match = re.match(pattern, temp)
        if match:
            temp = match.group(1) + "-" + match.group(2)

        return temp
    else:
        return None


# 遍历mp4
def scanDir(dir):
    dir = pathlib.Path(dir)
    return dir.glob("*.mp4")

# 要遍历的文件夹
origin_folder = filedialog.askdirectory(title="请选择一个文件夹")
print(origin_folder)
exit(0) if len(origin_folder) == 0 else 0

# 移动到目标文件夹
target_folder = r""

# 遍历
for i in scanDir(origin_folder):
    movie_id = recognize(i.stem)
    metadata = scrape(movie_id)
    if metadata:
        title = metadata["标题"]
        artist = metadata["演员"]
    else:
        continue

    try:
        try:
          # 先使用exiftool
            t = subprocess.run(
                f'exiftool "{i}" -Title="{title}" -Artist="{artist}" -overwrite_original -api largefilesupport=1',
                shell=True,
                check=True,
                encoding="utf-8",
                stdout=subprocess.PIPE,
            )
            shutil.move(i, target_folder)
            print("\033[0;32msuccess\n\033[0m")
        # 失败则使用ffmpeg
        except subprocess.CalledProcessError:
            print(f"try to use ffmpeg {i}")
            t = subprocess.run(
                f'ffmpeg -i "{i}" -metadata title="{title}" -metadata artist="{artist}" -c copy "{target_folder}\\{i.name}"',
                shell=True,
                check=True,
                encoding="utf-8",
                stdout=subprocess.PIPE,
            )
            os.remove(i)
            print("\033[0;32msuccess\n\033[0m")
    # 其他失败
    except Exception as e:
        print(f"\033[0;31merror {e}\n \033[0m")

# 检查有无遗漏
moives = []
for path, subdirs, files in os.walk(origin_folder):
    for name in files:
        filename = os.path.join(path, name)
        if not name.startswith(".") and os.path.splitext(filename)[1] in [".mp4", ".MP4"]:
            moives.append(filename)
if len(moives):
    pass
else:
    print("未处理的影片：")
    for moive in moives:
        print(moive)