#!/usr/bin/env python3
"""把新版 短剧库.zip 与本地源码对比，展示变更并更新本地旧代码。

用法:
    python scripts/update_from_zip.py                      # 从默认地址下载最新包并预览变更
    python scripts/update_from_zip.py --apply              # 下载并真正更新本地代码
    python scripts/update_from_zip.py 某个路径/短剧库.zip   # 用本地压缩包
    python scripts/update_from_zip.py --url https://...    # 指定其他下载地址

行为:
    1. 读取压缩包（自动处理中文文件名 GBK/UTF-8、自动剥离压缩包内公共顶层目录）
    2. 与本地源码逐文件比对（忽略 .git、data/、日志、压缩包等运行时文件）
    3. 列出 新增 / 修改 / 删除 三类变更；加 --apply 后写入本地
    4. 更新完成后用 git status / git diff 自查，再提交
"""
import argparse
import hashlib
import sys
import tempfile
import urllib.parse
import urllib.request
import zipfile
from pathlib import Path

DEFAULT_URL = "https://www.252035.xyz/短剧库.zip"
REPO_ROOT = Path(__file__).resolve().parent.parent
IGNORE_DIRS = {".git", "data", "__pycache__", "dist", "短剧下载"}
IGNORE_SUFFIX = {".log", ".zip", ".tar", ".pyc"}
IGNORE_NAMES = {".DS_Store", "Thumbs.db"}
PROTECT = {"scripts/update_from_zip.py"}  # 更新工具自身，永远不删


def download(url: str) -> Path:
    quoted = urllib.parse.quote(url, safe=":/?&=%")
    print(f"下载最新压缩包：{url}")
    request = urllib.request.Request(quoted, headers={"User-Agent": "juku-update/1.0"})
    with urllib.request.urlopen(request, timeout=60) as resp:
        total = int(resp.headers.get("Content-Length") or 0)
        target = Path(tempfile.gettempdir()) / "短剧库-update.zip"
        done = 0
        with target.open("wb") as f:
            while True:
                chunk = resp.read(1 << 16)
                if not chunk:
                    break
                f.write(chunk)
                done += len(chunk)
                if total:
                    print(f"\r  {done}/{total} 字节", end="", flush=True)
        print(f"\n  已保存到 {target}")
        return target


def norm_name(info: zipfile.ZipInfo) -> str:
    """修正 zip 内中文文件名。

    打包工具有两种常见错误：用 UTF-8 编码但不打 0x800 标志（跨平台工具），
    或用 GBK 编码且不打标志（中文 Windows 右键压缩）。依次尝试恢复：
    UTF-8 优先（其字节序列极少与 GBK 混淆），失败再试 GBK。
    """
    name = info.filename
    if not info.flag_bits & 0x800:
        try:
            raw = name.encode("cp437")
        except UnicodeEncodeError:
            return name.replace("\\", "/")
        for encoding in ("utf-8", "gbk"):
            try:
                name = raw.decode(encoding)
                break
            except UnicodeDecodeError:
                continue
    return name.replace("\\", "/")


def ignored(rel: str) -> bool:
    parts = rel.split("/")
    return (
        any(p in IGNORE_DIRS for p in parts[:-1])
        or Path(rel).name in IGNORE_NAMES
        or Path(rel).suffix in IGNORE_SUFFIX
    )


def common_prefix(names: list[str]) -> str:
    """所有条目共享同一顶层目录时返回该前缀（如 '短剧库/'），否则返回空串。"""
    if not names:
        return ""
    tops = {n.split("/", 1)[0] for n in names}
    if len(tops) == 1 and all("/" in n for n in names):
        return next(iter(tops)) + "/"
    return ""


def md5_of(path: Path) -> str:
    h = hashlib.md5()
    with path.open("rb") as f:
        for chunk in iter(lambda: f.read(1 << 16), b""):
            h.update(chunk)
    return h.hexdigest()


def main() -> int:
    parser = argparse.ArgumentParser(description="用新版 zip 更新本地短剧库源码")
    parser.add_argument("zipfile", nargs="?", default=None,
                        help="本地压缩包路径；省略则从 --url 地址下载最新包")
    parser.add_argument("--url", default=DEFAULT_URL, help=f"新版下载地址（默认 {DEFAULT_URL}）")
    parser.add_argument("--apply", action="store_true", help="真正写入本地（默认只预览）")
    args = parser.parse_args()

    if args.zipfile:
        zip_path = Path(args.zipfile)
        if not zip_path.exists():
            print(f"找不到压缩包：{args.zipfile}")
            return 1
    else:
        zip_path = download(args.url)

    with zipfile.ZipFile(zip_path) as zf:
        infos = [i for i in zf.infolist() if not i.is_dir()]
        names = [norm_name(i) for i in infos]
        # 先丢弃 macOS 压缩附带的资源文件，避免干扰顶层目录识别
        pairs = [
            (info, name)
            for info, name in zip(infos, names)
            if "__MACOSX" not in name.split("/") and not Path(name).name.startswith("._")
        ]
        prefix = common_prefix([name for _, name in pairs])
        if prefix:
            print(f"检测到压缩包顶层目录：{prefix}（比对时已忽略）")
        incoming: dict[str, bytes] = {}
        for info, name in pairs:
            rel = name[len(prefix):] if prefix else name
            if rel and not ignored(rel):
                incoming[rel] = zf.read(info)

    added, modified, unchanged = [], [], []
    for rel, blob in sorted(incoming.items()):
        local = REPO_ROOT / rel
        if not local.exists():
            added.append(rel)
        elif md5_of(local) != hashlib.md5(blob).hexdigest():
            modified.append(rel)
        else:
            unchanged.append(rel)

    local_files: set[str] = set()
    for path in REPO_ROOT.rglob("*"):
        if not path.is_file():
            continue
        rel = path.relative_to(REPO_ROOT).as_posix()
        if ignored(rel) or rel in IGNORE_NAMES:
            continue
        local_files.add(rel)
    deleted = sorted(local_files - set(incoming) - PROTECT)

    print(f"\n对比完成：压缩包 {len(incoming)} 个文件，本地 {len(local_files)} 个文件")
    print(f"  新增 {len(added)} 个")
    for rel in added:
        print(f"    + {rel}")
    print(f"  修改 {len(modified)} 个")
    for rel in modified:
        print(f"    ~ {rel}")
    print(f"  删除 {len(deleted)} 个")
    for rel in deleted:
        print(f"    - {rel}")
    print(f"  未变化 {len(unchanged)} 个")

    if not (added or modified or deleted):
        print("本地代码与压缩包一致，无需更新。")
        return 0
    if not args.apply:
        print("\n当前为预览模式。确认无误后追加 --apply 执行更新，然后 git add -A && git commit。")
        return 0

    for rel in added + modified:
        target = REPO_ROOT / rel
        target.parent.mkdir(parents=True, exist_ok=True)
        target.write_bytes(incoming[rel])
        print(f"已写入  {rel}")
    for rel in deleted:
        (REPO_ROOT / rel).unlink(missing_ok=True)
        print(f"已删除  {rel}")
    print("\n更新完成。请执行 git status / git diff 复核后提交。")
    return 0


if __name__ == "__main__":
    sys.exit(main())