import json
import os
from typing import Dict, List

import httpx

from config import Config


def load_templates(dir_path: str) -> List[Dict[str, any]]:
    items = []
    for name in os.listdir(dir_path):
        if not name.endswith(".json"):
            continue
        with open(os.path.join(dir_path, name), "r", encoding="utf-8") as f:
            items.append(json.load(f))
    return items


def build_document(item: Dict[str, any]) -> str:
    table = item.get("table", "")
    templates = item.get("templates", [])
    lines = [f"table: {table}"]
    for t in templates:
        lines.append(f"- {t}")
    return "\n".join(lines)


def main():
    cfg = Config.get_section("ragflow")
    base_url = cfg.get("url", "").rstrip("/")
    api_key = cfg.get("api_key", "")
    dataset_ids = cfg.get("dataset_ids", "")

    upload_path = cfg.get("upload", "")
    if not upload_path:
        raise RuntimeError("未配置 ragflow.upload 上传接口，请在 config/config.ini 中设置")

    url = base_url + "/" + upload_path.lstrip("/")
    headers = {"Authorization": f"Bearer {api_key}"} if api_key else {}

    root = os.path.dirname(os.path.dirname(__file__))
    templates_dir = os.path.join(root, "config", "ragflow_templates")
    items = load_templates(templates_dir)

    payload = []
    for item in items:
        payload.append(
            {
                "dataset_ids": dataset_ids,
                "content": build_document(item),
                "metadata": {"table": item.get("table")},
            }
        )

    with httpx.Client(timeout=60) as client:
        response = client.post(url, json={"documents": payload}, headers=headers)
    response.raise_for_status()
    print("同步完成")


if __name__ == "__main__":
    main()