From 8255070f60c8c055f0f6ae8d653426a336460a7b Mon Sep 17 00:00:00 2001 From: zy123 <646228430@qq.com> Date: Thu, 17 Oct 2024 16:47:35 +0800 Subject: [PATCH] =?UTF-8?q?10.17=20=E5=B0=8F=E8=A7=A3=E6=9E=90=E8=B4=A7?= =?UTF-8?q?=E7=89=A9=E6=A0=87?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- flask_app/general/little_zbparse.py | 49 ++++++++++++++++++----------- flask_app/main/start_up.py | 6 ++-- 2 files changed, 33 insertions(+), 22 deletions(-) diff --git a/flask_app/general/little_zbparse.py b/flask_app/general/little_zbparse.py index 86e11d3..665a993 100644 --- a/flask_app/general/little_zbparse.py +++ b/flask_app/general/little_zbparse.py @@ -22,34 +22,45 @@ def get_global_logger(unique_id): logger = None -def get_base_info(baseinfo_file_path): + +def merge(merged): + guarantee_key = '是否递交投标保证金' if '是否递交投标保证金' in merged else '是否递交磋商保证金' + if merged.get(guarantee_key) == '是': + return 1,"" + elif merged.get(guarantee_key) == '否': + guarantee_type = '投标' if '投标' in guarantee_key else '磋商' + merged[f'{guarantee_type}保证金'] = '不提交' + merged[f'退还{guarantee_type}保证金'] = '/' + merged.pop(guarantee_key, None) + return 0,merged + +def get_goods_baseinfo(baseinfo_file_path): file_id = upload_file(baseinfo_file_path) - baseinfo_file_path='flask_app/static/提示词/基本信息货物标.txt' - # baseinfo_file_path = 'D:\\flask_project\\flask_app\\static\\提示词\\基本信息货物标.txt' + # baseinfo_file_path='flask_app/static/提示词/基本信息货物标.txt' + baseinfo_file_path = 'D:\\flask_project\\flask_app\\static\\提示词\\基本信息货物标.txt' questions = read_questions_from_file(baseinfo_file_path) - more_query = "请你根据招标文件信息，回答以下问题：是否组织踏勘现场？是否召开投标预备会（或投标答疑会）？是否退还投标文件？是否允许分包? 是否需要递交投标保证金（或磋商保证金）？是否需要提交履约保证金（或履约担保）？是否有招标代理服务费（或中标、成交服务费）？请按json格式给我提供信息，键名分别为'是否组织踏勘现场','是否召开投标预备会'（或'是否召开投标答疑会'）,'是否退还投标文件',是否允许分包','是否递交投标保证金'（或'是否递交磋商保证金'）,'是否提交履约保证金','是否有招标代理服务费',键值仅限于'是','否','未知',若存在矛盾信息，请回答'未知'。" + more_query = "请你根据招标文件信息，回答以下问题：是否需要递交投标保证金（或磋商保证金）？请按json格式给我提供信息，键名分为'是否递交投标保证金'（或'是否递交磋商保证金'）,键值仅限于'是','否','未知',若存在矛盾信息，请回答'未知'。" questions.append(more_query) baseinfo_results = multi_threading(questions, "", file_id, 2) # 1代表使用百炼rag 2代表使用qianwen-long baseinfo_list = [clean_json_string(res) for _, res in baseinfo_results] if baseinfo_results else [] - chosen_numbers, merged = merge_json_to_list(baseinfo_list.pop()) - baseinfo_list.append(merged) - - judge_file_path = 'flask_app/static/提示词/是否相关问题货物标.txt' - # judge_file_path = 'D:\\flask_project\\flask_app\\static\\提示词\\是否相关问题货物标.txt' - judge_questions = read_questions_from_judge(judge_file_path, chosen_numbers) - - res2 = multi_threading(judge_questions, "", file_id, 2) # 调用千问-long - if not res2: - print("基础信息整合： multi_threading error!") + type,merged=merge(baseinfo_list.pop()) + if type: + judge_questions="根据招标文件第二章投标人须知，该项目投标保证金（或磋商保证金）的内容或要求是什么？请按json格式给我提供信息，外层键名为'投标保证金'（或'磋商保证金'），若需要以嵌套键值对返回结果，那么嵌套键名为你对相应要求的总结，而对应键值需要完全与原文保持一致。" + res2 = multi_threading(judge_questions, "", file_id, 2) # 调用千问-long + if not res2: + print("基础信息整合： multi_threading error!") + else: + for question, response in res2: + baseinfo_list.append(clean_json_string(response)) else: - for question, response in res2: - baseinfo_list.append(clean_json_string(response)) + baseinfo_list.append(merged) return baseinfo_list + #货物标 def little_parse_goods(output_folder,file_path): files=truncate_pdf_specific(file_path,output_folder) - baseinfo_list=get_base_info(files[-1]) + baseinfo_list=get_goods_baseinfo(files[-1]) aggregated_baseinfo = aggregate_basic_info_goods(baseinfo_list) return {"基础信息": aggregated_baseinfo} def little_parse_engineering(output_folder,downloaded_filepath): @@ -107,8 +118,8 @@ if __name__ == "__main__": start_time = time.time() file_type = 2 # 1:docx 2:pdf 3:其他 zb_type=2 #1:工程标 2：货物标 - input_file = "C:\\Users\\Administrator\\Desktop\\货物标\\zbfiles\\6.2定版视频会议磋商文件.pdf" - res=little_parse_main(output_folder, input_file, file_type, zb_type) + input_file = "C:\\Users\\Administrator\\Desktop\\货物标\\zbfiles\\ztbfile.pdf" + res=little_parse_main(output_folder, input_file, file_type, zb_type,"122334") print(json.dumps(res, ensure_ascii=False, indent=4)) end_time = time.time() elapsed_time = end_time - start_time # 计算耗时 diff --git a/flask_app/main/start_up.py b/flask_app/main/start_up.py index 0c84643..487e81a 100644 --- a/flask_app/main/start_up.py +++ b/flask_app/main/start_up.py @@ -295,15 +295,15 @@ def process_and_stream(file_url, zb_type): final_result, extracted_info = outer_post_processing(combined_data, includes) try: - with open(output_json_path, 'w', encoding='utf-8') as json_file: + with open(extracted_info_path, 'w', encoding='utf-8') as json_file: json.dump(extracted_info, json_file, ensure_ascii=False, indent=4) - logger.info(f"合并后的数据已保存到 '{output_json_path}'") + logger.info(f"摘取后的数据已保存到 '{extracted_info_path}'") except IOError as e: logger.error(f"保存JSON文件时出错: {e}") try: with open(output_json_path, 'w', encoding='utf-8') as json_file: - json.dump(extracted_info_path, json_file, ensure_ascii=False, indent=4) + json.dump(final_result, json_file, ensure_ascii=False, indent=4) logger.info(f"合并后的数据已保存到 '{output_json_path}'") except IOError as e: logger.error(f"保存JSON文件时出错: {e}")