import json

from flask_app.main.通义千问long import upload_file, qianwen_long
from flask_app.main.json_utils import clean_json_string

def combine_technical_and_business(data, target_values1, target_values2):
    extracted_data = {}  # 根级别存储所有数据
    technical_found = False
    business_found = False

    def extract_nested(data, parent_key='', is_technical=False, is_business=False):
        nonlocal technical_found, business_found
        if isinstance(data, dict):
            for key, value in data.items():
                current_key = f"{parent_key}.{key}" if parent_key else key

                # 检查是否为技术标的内容
                if any(target in key for target in target_values1):     #模糊匹配
                    if not is_technical:
                        # 直接存储在根级别
                        extracted_data[key] = value
                        technical_found = True
                        # 标记为技术标内容并停止进一步处理这个分支
                        continue

                # 检查是否为商务标的内容
                elif any(target in key for target in target_values2):
                    if not is_business:
                        # 存储在'商务标'分类下
                        if '商务标' not in extracted_data:
                            extracted_data['商务标'] = {}
                        extracted_data['商务标'][key] = value
                        business_found = True
                        # 标记为商务标内容并停止进一步处理这个分支
                        continue

                # 如果当前值是字典或列表，且不在技术或商务分类下，继续递归搜索
                if isinstance(value, dict) or isinstance(value, list):
                    extract_nested(value, current_key, is_technical, is_business)

        elif isinstance(data, list):
            for index, item in enumerate(data):
                extract_nested(item, f"{parent_key}[{index}]", is_technical, is_business)

    # 开始从顶级递归搜索
    extract_nested(data)

    # 处理未找到匹配的情况
    if not technical_found:
        extracted_data['技术标'] = ''
    if not business_found:
        extracted_data['商务标'] = ''

    return extracted_data


def get_evaluation_standards(truncate_file):
    file_id = upload_file(truncate_file)
    user_query = "根据该文档中的评标办法前附表，请你列出该文件的技术标，商务标，投标报价评审标准以及它们对应的具体评分要求，若对应内容中存在其他信息，在键名如'技术标'中新增子键名'备注'存放该信息。如果评分内容不是这3个，则返回文档中给定的评分内容以及它的评分要求，都以json的格式返回结果。请不要回答有关形式、资格、响应性评审标准的内容"
    evaluation_res = qianwen_long(file_id, user_query)
    cleaned_evaluation_res = clean_json_string(evaluation_res)

    include = ['一包', '二包', '三包', '四包', '五包']
    target_values1 = ['技术标', '技术部分', '设计', '实施', '方案']
    target_values2 = ['投标报价', '商务标', '商务部分', '报价部分', '业绩', '信誉', '分值', '计算公式', '信用', '人员',
                      '资格', '奖项', '认证', '荣誉']
    updated_jsons = {}

    for key in cleaned_evaluation_res.keys():
        if any(item in key for item in include):
            inner_dict = cleaned_evaluation_res[key]
            # 将处理后的结果存储到updated_jsons中，每个包名为键
            updated_jsons[key] = combine_technical_and_business(inner_dict, target_values1, target_values2)

    # 将updated_jsons转换为JSON格式
    evaluation_combined_res = json.dumps(updated_jsons, ensure_ascii=False, indent=4)
    return evaluation_combined_res
def combine_technical_and_business(data, target_values1, target_values2):
    extracted_data = {}  # 根级别存储所有数据
    technical_found = False
    business_found = False

    def extract_nested(data, parent_key='', is_technical=False, is_business=False):
        nonlocal technical_found, business_found
        if isinstance(data, dict):
            for key, value in data.items():
                current_key = f"{parent_key}.{key}" if parent_key else key

                # 检查是否为技术标的内容
                if any(target in key for target in target_values1):
                    if not is_technical:
                        # 直接存储在根级别
                        extracted_data[key] = value
                        technical_found = True
                        # 标记为技术标内容并停止进一步处理这个分支
                        continue

                # 检查是否为商务标的内容
                elif any(target in key for target in target_values2):
                    if not is_business:
                        # 存储在'商务标'分类下
                        if '商务标' not in extracted_data:
                            extracted_data['商务标'] = {}
                        extracted_data['商务标'][key] = value
                        business_found = True
                        # 标记为商务标内容并停止进一步处理这个分支
                        continue

                # 如果当前值是字典或列表，且不在技术或商务分类下，继续递归搜索
                if isinstance(value, dict) or isinstance(value, list):
                    extract_nested(value, current_key, is_technical, is_business)

        elif isinstance(data, list):
            for index, item in enumerate(data):
                extract_nested(item, f"{parent_key}[{index}]", is_technical, is_business)

    # 开始从顶级递归搜索
    extract_nested(data)

    # 处理未找到匹配的情况
    if not technical_found:
        extracted_data['技术标'] = ''
    if not business_found:
        extracted_data['商务标'] = ''

    return extracted_data

if __name__ == "__main__":
    truncate_file="C:\\Users\\Administrator\\Desktop\\货物标\\output2\\招标文件(107国道)_evaluation_method.pdf"
    res=get_evaluation_standards(truncate_file)
    cleaned_res=clean_json_string(res).get("")