zbparse/flask_app/货物标/提取采购需求main.py

import concurrent.futures
import json
import time
from flask_app.货物标.技术参数要求提取 import get_technical_requirements
from flask_app.general.通义千问long import upload_file
from flask_app.货物标.商务服务其他要求提取 import get_business_requirements


#获取采购清单
def fetch_procurement_reqs(procurement_path,procurement_docpath):
    # 定义默认的 procurement_reqs 字典
    DEFAULT_PROCUREMENT_REQS = {
        "技术要求": "",
        "商务要求": "",
        "服务要求": "",
        "其他要求": ""
    }
    # 如果 truncate_file 是空字符串，直接返回包含空字符串的字典
    if not procurement_docpath:
        return DEFAULT_PROCUREMENT_REQS.copy()
    try:
        # 上传文件并获取 file_id
        file_id = upload_file(procurement_docpath)

        # 使用 ThreadPoolExecutor 并行处理 get_technical_requirements 和 get_business_requirements
        with concurrent.futures.ThreadPoolExecutor() as executor:
            # 提交任务给线程池
            future_technical = executor.submit(get_technical_requirements, file_id)
            time.sleep(1)  # 如果需要延迟，可以保留，否则建议移除以提高效率
            future_business = executor.submit(get_business_requirements, procurement_path, file_id)

            # 获取并行任务的结果
            technical_requirements = future_technical.result()
            business_requirements = future_business.result()

        # 构建最终的嵌套结构，确保四个键平级
        procurement_reqs = {
            "技术要求": technical_requirements.get("技术要求", {}),
            "商务要求": business_requirements.get("商务要求", {}),
            "服务要求": business_requirements.get("服务要求", {}),
            "其他要求": business_requirements.get("其他要求", {}),
        }

        return procurement_reqs

    except Exception as e:
        print(f"Error in fetch_procurement_reqs: {e}")
        # 在出错时返回默认的包含空字符串的字典
        return DEFAULT_PROCUREMENT_REQS.copy()

if __name__ == "__main__":
    output_folder = "C:\\Users\\Administrator\\Desktop\\货物标\\货物标output"
    # file_path="C:\\Users\\Administrator\\Desktop\\货物标\\output1\\2-招标文件（2020年广水市中小学教师办公电脑系统及多媒体“班班通”设备采购安装项目）_procurement.pdf"
    file_path="D:\\flask_project\\flask_app\\static\\output\\output1\\bf225a5e-16d0-45c8-8c19-54a1a94cf3e2\\ztbfile_procurement.docx"
    res=fetch_procurement_reqs(file_path)
    print(json.dumps(res, ensure_ascii=False, indent=4))