"""将 backend/data/books/*.json 导入 word_books / word_book_entries 表。"""

from __future__ import annotations

import json
import sys
from pathlib import Path

BACKEND_DIR = Path(__file__).resolve().parents[1]
if str(BACKEND_DIR) not in sys.path:
    sys.path.insert(0, str(BACKEND_DIR))

from database import SessionLocal  # noqa: E402
from models import WordBook, WordBookEntry  # noqa: E402
from services.book_service import DEPRECATED_SLUGS  # noqa: E402

BOOKS_DIR = BACKEND_DIR / "data" / "books"


def upsert_book(session, data: dict) -> WordBook:
    book = session.query(WordBook).filter(WordBook.slug == data["slug"]).first()
    if not book:
        book = WordBook(slug=data["slug"])
        session.add(book)
    book.title = data["title"]
    book.description = data.get("description")
    book.level = data.get("level", "general")
    book.word_count = data.get("word_count", 0)
    book.unit_count = data.get("unit_count", 0)
    book.sort_order = data.get("sort_order", 0)
    book.daily_target = data.get("daily_target", 15)
    book.master_required_count = data.get("master_required_count", 4)
    book.weak_wrong_threshold = data.get("weak_wrong_threshold", 2)
    book.learn_mode = data.get("learn_mode", "sequential")
    book.is_published = 1
    session.flush()
    return book


def import_book_file(session, path: Path) -> tuple[str, int]:
    data = json.loads(path.read_text(encoding="utf-8"))
    book = upsert_book(session, data)
    session.query(WordBookEntry).filter(WordBookEntry.book_id == book.id).delete(
        synchronize_session=False
    )
    count = 0
    for unit in data.get("units", []):
        unit_no = unit.get("unit", 1)
        unit_title = unit.get("title")
        for w in unit.get("words", []):
            entry = WordBookEntry(
                book_id=book.id,
                unit=unit_no,
                unit_title=unit_title,
                sort_index=w["sort_index"],
                lemma_en=w["lemma_en"].strip().lower()[:128],
                zh=w["zh"].strip()[:256],
                phonetic=(w.get("phonetic") or "")[:128] or None,
                example_en=w.get("example_en"),
                example_cn=w.get("example_cn"),
            )
            session.add(entry)
            count += 1
    book.word_count = count
    book.unit_count = len(data.get("units", []))
    session.commit()
    return book.slug, count


def import_all(books_dir: Path = BOOKS_DIR) -> None:
    files = sorted(books_dir.glob("*.json"))
    files = [
        f
        for f in files
        if f.name != "manifest.json"
        and f.stem not in DEPRECATED_SLUGS
    ]
    if not files:
        print(f"未找到词书 JSON: {books_dir}")
        return
    session = SessionLocal()
    try:
        for path in files:
            slug, count = import_book_file(session, path)
            print(f"  导入 {slug}: {count} 词条")
        for slug in DEPRECATED_SLUGS:
            book = session.query(WordBook).filter(WordBook.slug == slug).first()
            if book and book.is_published:
                book.is_published = 0
                session.commit()
                print(f"  下架旧词书 {slug}")
    finally:
        session.close()


def main() -> None:
    import argparse

    parser = argparse.ArgumentParser(description="导入词书 JSON 到数据库")
    parser.add_argument("--dir", type=Path, default=BOOKS_DIR)
    args = parser.parse_args()
    import_all(args.dir)


if __name__ == "__main__":
    main()