# -*- coding: utf-8 -*-
"""从美团「城市字典」Excel 生成随仓库的 JSON(app/integrations/data/meituan_cities.json)。

本地一次性工具:美团每年更新城市字典时,把新 Excel 放进来重跑即可。
Excel 不入库、不进仓库(二进制、需 openpyxl);生成的 JSON 随仓库提交,部署到服务器后
ETL(scripts/pull_meituan_coupons.py)直接读它遍历全部城市。

字典口径:359 个地级市,经实测一个地级市 cityId 已覆盖其下辖县级市(如徐州→邳州/新沂),
故无需区县层级。

用法:
    python tools/gen_meituan_cities.py ["城市字典xxx.xlsx" 路径]
    (不传则用默认 e:\\codes\\城市字典2025 (1).xlsx)
"""
from __future__ import annotations

import json
import os
import sys

import openpyxl

HERE = os.path.dirname(os.path.abspath(__file__))
ROOT = os.path.dirname(HERE)
DEFAULT_EXCEL = r"e:\codes\城市字典2025 (1).xlsx"
OUT = os.path.join(ROOT, "app", "integrations", "data", "meituan_cities.json")


def main() -> None:
    excel = sys.argv[1] if len(sys.argv) > 1 else DEFAULT_EXCEL
    wb = openpyxl.load_workbook(excel, data_only=True)
    ws = wb.worksheets[0]
    rows = list(ws.iter_rows(values_only=True))
    header = rows[0]

    cities = []
    seen = set()
    for r in rows[1:]:
        cid = str(r[0]).strip() if r[0] else ""
        name = str(r[1]).strip() if len(r) > 1 and r[1] else ""
        prov = str(r[2]).strip() if len(r) > 2 and r[2] else ""
        if not cid or cid in seen:
            continue
        seen.add(cid)
        cities.append({"city_id": cid, "name": name, "province": prov})

    os.makedirs(os.path.dirname(OUT), exist_ok=True)
    with open(OUT, "w", encoding="utf-8") as f:
        json.dump(cities, f, ensure_ascii=False, indent=1)

    print(f"表头: {header}")
    print(f"写出 {len(cities)} 城 → {OUT}")
    print("样例:", cities[:3])


if __name__ == "__main__":
    main()