cli.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
统一 CLI：豆瓣读书/影视迁移工具

用法：
    python3 cli.py --book               # 爬全部 + 生成 Goodreads CSV
    python3 cli.py --movie              # 爬全部 + 生成 Letterboxd CSV
    python3 cli.py --book --movie      # 两者同时执行
    python3 cli.py --book --mode test  # 测试模式
    python3 cli.py --book --mode limited  # 有限模式
    python3 cli.py --help
"""
import sys
import argparse
import os
import logging

import shared
import books
import movies

logger = logging.getLogger("cli")


# ==================== CLI 参数解析 ====================

def build_parser():
    parser = argparse.ArgumentParser(
        description="豆瓣读书/影视迁移工具",
        formatter_class=argparse.RawDescriptionHelpFormatter,
        epilog=(
            "\n"
            "示例：\n"
            "  python3 cli.py --book                       # 爬取全部图书\n"
            "  python3 cli.py --movie                     # 爬取全部影视\n"
            "  python3 cli.py --book --movie            # 同时爬取两者\n"
            "  python3 cli.py --book --mode test       # 测试模式（极小量）\n"
            "  python3 cli.py --book --mode limited     # 有限模式（2 页）\n"
            "  python3 cli.py --book --dump-html        # 保存 HTML 供调试\n"
            "  python3 cli.py --book --no-import-review  # 不输出评论\n"
        ),
    )

    parser.add_argument(
        "--book", dest="target", action="append_const", const="book",
        help="爬取豆瓣图书 → Goodreads CSV",
    )
    parser.add_argument(
        "--movie", dest="target", action="append_const", const="movie",
        help="爬取豆瓣影视 → Letterboxd CSV",
    )

    parser.add_argument(
        "--mode", "-m",
        choices=["test", "limited", "full"],
        default="full",
        help="运行模式：test=极小测试，limited=有限运行，full=完整运行（默认）",
    )

    parser.add_argument(
        "--pages", "-p", type=int, default=None,
        help="强制设置最大爬取页数",
    )

    parser.add_argument(
        "--resume", "-r", action="store_true",
        help="从上次断点继续",
    )

    parser.add_argument(
        "--dump-html", action="store_true",
        help="将所有响应 HTML 保存到 output/.work/html_dump/ 供调试",
    )

    parser.add_argument(
        "--ua", default=None,
        help="覆盖默认 User-Agent",
    )

    parser.add_argument(
        "--no-import-review", dest="import_review", action="store_false",
        help="不在 CSV 中包含评论（默认输出评论）",
    )
    parser.set_defaults(import_review=True)

    return parser


# ==================== 辅助 ====================

def _resolve_max_pages(mode, explicit_pages):
    if explicit_pages is not None:
        return explicit_pages
    if mode == "test":
        return 1
    if mode == "limited":
        return 2
    return None  # full


def _apply_ua_override(ua):
    if ua:
        shared.DEFAULT_UA = ua
        logger.info(f"User-Agent 已覆盖：{ua}")


def _test_auth(target):
    from shared import request_with_retry, get_default_headers
    if target == "book":
        from books import BASE_URL as B
        from shared import USER_ID as U
        url = f"{B}/people/{U}/collect"
    else:
        from movies import BASE_URL as M
        from shared import USER_ID as U
        url = f"{M}/people/{U}/collect"
    logger.info(f"正在验证 Cookie（{target}）...")
    resp = request_with_retry("GET", url, headers=get_default_headers())
    if resp:
        logger.info(f"Cookie 验证成功，返回状态码：{resp.status_code}")


# ==================== Book ====================

def run_book(args):
    max_pages = _resolve_max_pages(args.mode, args.pages)
    import_review = args.import_review

    if args.mode == "test":
        _test_auth("book")
        logger.info("[test] 爬取图书列表第 1 页（无延迟）...")
        result = books.scrape(start_page=0, max_pages=1)
        if result:
            url = result[0]["url"]
            title = result[0]["title"]
            isbn, detail_title = books._fetch_book_detail(url)
            logger.info(f"[test] 《{detail_title or title}》ISBN: {isbn}")
        logger.info("[test] 图书测试完成！")
        return

    if args.resume:
        start = books._load_checkpoint()
        if start == 0:
            logger.info("未发现有效断点，将从头开始")
    else:
        books._clear_checkpoint()
        start = 0

    logger.info(f"开始爬取图书列表（最多 {max_pages or '全部'} 页）...")
    result = books.scrape(start_page=start, max_pages=max_pages)

    if not result:
        logger.error("图书列表爬取失败")
        sys.exit(1)

    if args.resume and os.path.exists(shared.CHECKPOINT_ISBN):
        logger.info("从断点继续 Goodreads CSV 生成...")
        books.build_goodreads_csv(resume=True, import_review=import_review)
    elif args.mode == "limited":
        n = (max_pages or 2) * 15
        books.build_goodreads_csv(records=result[:n], import_review=import_review)
    else:
        books.build_goodreads_csv(records=result, import_review=import_review)

    logger.info("图书迁移全部完成！")


# ==================== Movie ====================

def run_movie(args):
    max_pages = _resolve_max_pages(args.mode, args.pages)
    import_review = args.import_review

    if args.mode == "test":
        _test_auth("movie")
        logger.info("[test] 爬取影视列表第 1 页（无延迟）...")
        result = movies.scrape(start_page=0, max_pages=1)
        if result:
            url = result[0]["url"]
            title = result[0]["title"]
            imdb_id, directors, year, orig_title = movies._fetch_letterboxd_fields(url)
            logger.info(f"[test] 《{title}》→ {orig_title} | IMDB: {imdb_id} | 导演: {directors[:40]} | Year: {year}")
        logger.info("[test] 影视测试完成！")
        return

    if args.resume:
        start = movies._load_checkpoint()
        if start == 0:
            logger.info("未发现有效断点，将从头开始")
    else:
        movies._clear_checkpoint()
        start = 0

    logger.info(f"开始爬取影视列表（最多 {max_pages or '全部'} 页）...")
    result = movies.scrape(start_page=start, max_pages=max_pages)

    if not result:
        logger.error("影视列表爬取失败")
        sys.exit(1)

    if args.resume and os.path.exists(shared.CHECKPOINT_LETTERBOXD):
        logger.info("从断点继续 Letterboxd CSV 生成...")
        movies.build_letterboxd_csv(resume=True, import_review=import_review)
    elif args.mode == "limited":
        n = (max_pages or 2) * 15
        movies.build_letterboxd_csv(records=result[:n], import_review=import_review)
    else:
        movies.build_letterboxd_csv(records=result, import_review=import_review)

    logger.info("影视迁移全部完成！")


# ==================== 主入口 ====================

def main():
    parser = build_parser()
    args = parser.parse_args()

    if not args.target:
        parser.print_help()
        print()
        logger.info("请使用 --book 或 --movie 指定要迁移的数据类型。")
        logger.info("例如：python3 cli.py --book")
        sys.exit(0)

    if args.dump_html:
        shared.DUMP_HTML = True
        logger.info("HTML 转储已开启，内容将保存至 output/.work/html_dump/")

    if args.ua:
        _apply_ua_override(args.ua)

    for t in args.target:
        logger.info("=" * 60)
        label = "图书（→ Goodreads）" if t == "book" else "影视（→ Letterboxd）"
        logger.info(f"开始执行：{label}")
        logger.info("=" * 60)
        if t == "book":
            run_book(args)
        elif t == "movie":
            run_movie(args)


if __name__ == "__main__":
    main()