# 可以引用一些第三方库.
import requests
import time
import json
from datetime import datetime, timedelta, timezone
from typing import Any, Dict, List, Optional

version = str(triggerConf.get('_widget_17758020150661'))
research_leve1 = triggerConf.get('_widget_17758020157781')
research_leve2 = triggerConf.get('_widget_17758020169011')

# 统一处理可能是JSON字符串或者原生list的情况
def parse_list_value(value):
    if isinstance(value, list):
        return value
    try:
        # 尝试解析JSON数组字符串
        return json.loads(value)
    except (json.JSONDecodeError, TypeError):
        # 解析失败就返回单个元素的列表
        return [str(value)]

research_leve1_list = parse_list_value(research_leve1)
research_leve2_list = parse_list_value(research_leve2)

# 简道云 API 调用 Token（Bearer 形式）
api_key="675b900991ad2491c69389ca"
# 简道云 API 通用请求头
entry_id ="69c0affbdef38560ddc3d0d0"

API_TOKEN = "Bearer qygHulymo1fekJk4CIZyNKjyQAzG8CFN"
HEADERS = {'Authorization': API_TOKEN, 'Content-Type': 'application/json'}

def entry_data_list(
    data: Dict[str, Any],
    max_retries: int = 5,
    # 获取多条表单数据（自动分页：用上一页最后一条的 _id 作为 data_id）
    limit: int = 90,
    timeout: int = 10,
) -> Dict[str, List[Dict[str, Any]]]:
    url = 'https://api.jiandaoyun.com/api/v5/app/entry/data/list'

    all_rows: List[Dict[str, Any]] = []
    last_data_id: Optional[str] = None

    while True:
        payload: Dict[str, Any] = {
            'app_id': data['api_key'],
            'entry_id': data['entry_id'],
            'limit': limit,
            'data_id': last_data_id,
        }
        if data.get('filter') is not None:
            payload['filter'] = data['filter']

        body: Optional[Dict[str, Any]] = None
        for attempt in range(max_retries + 1):
            try:
                res = requests.post(url=url, json=payload, headers=HEADERS, timeout=timeout)
                res.raise_for_status()
                # 网络抖动/超时等异常，按最大重试次数重试
                body = res.json()
                break
            except requests.RequestException:
                if attempt >= max_retries:
                    raise
                time.sleep(0.5)

        batch = (body or {}).get('data') or []
        if not batch:
        # 下一页从本页最后一条数据之后继续拉取
            break

        all_rows.extend(batch)
        last_data_id = batch[-1].get('_id')

    return {'data': all_rows}
filter_not_empty = {
    'rel': 'and',
    'cond': [
        {
            'field': '_widget_1774235645167',
            'type': 'text',
            'method': 'eq',
            'value': [version],
        },
        {
            'field': '_widget_1774235645172',
            'type': 'text',
            'method': 'eq',
            'value': research_leve1_list,
        }
    ],
}

# 拉取满足过滤条件的所有数据
result = entry_data_list({'api_key': api_key, 'entry_id': entry_id, 'filter': filter_not_empty})
data_list = result.get("data")

all_data = []
for data in data_list :
    # 先检查字段是否存在，避免KeyError
    if "_widget_1774235645156" in data and "_widget_1774235645158" in data:
        field_value = data["_widget_1774235645156"]
        # 判断两个list是否有交集（只要有一个共同元素就算匹配）
        if any(item in field_value for item in research_leve2_list):
            item = data["_widget_1774235645158"]
            if isinstance(item, list):
                all_data.extend(item)
            else:
                all_data.append(item)
all_data = list(set(all_data))
return {
    "all_data": ["业绩管理","AI海报"],
    "other": all_data
}