12.12 解决了clause_path为空 开评定标失败的bug
This commit is contained in:
parent
c77ada480c
commit
528c4b38a1
@ -423,13 +423,13 @@ def combine_evaluation_standards(evaluation_method_path,invalid_path,zb_type):
|
||||
if __name__ == "__main__":
|
||||
start_time=time.time()
|
||||
# truncate_file=r"C:\Users\Administrator\Desktop\招标文件-采购类\tmp2\2024-新疆-塔城地区公安局食药环分局快检实验室项目_evaluation_method.pdf"
|
||||
evaluation_method_path = r'C:\Users\Administrator\Desktop\fsdownload\91399aa4-1ee8-447d-a05b-03cd8d15ced5\ztbfile_evaluation_method.pdf'
|
||||
evaluation_method_path = r'C:\Users\Administrator\Desktop\fsdownload\aba81749-5986-4492-8b4b-16db9c69a09d\ztbfile_evaluation_method.pdf'
|
||||
invalid_path=r'C:\Users\Administrator\Desktop\fsdownload\91399aa4-1ee8-447d-a05b-03cd8d15ced5\ztbfile_invalid.pdf'
|
||||
# truncate_file = "C:\\Users\\Administrator\\Desktop\\货物标\\output2\\2-招标文件（统计局智能终端二次招标）_evaluation_method.pdf"
|
||||
# truncate_file="C:\\Users\\Administrator\\Desktop\\货物标\\output2\\广水市妇幼招标文件最新（W改）_evaluation_method.pdf"
|
||||
# truncate_file = "C:\\Users\\Administrator\\Desktop\\fsdownload\\2d481945-1f82-45a5-8e56-7fafea4a7793\\ztbfile_evaluation_method.pdf"
|
||||
# truncate_file="C:\\Users\\Administrator\\Desktop\\fsdownload\\ztbfile_evaluation_method.pdf"
|
||||
res = combine_evaluation_standards(evaluation_method_path,invalid_path,1)
|
||||
res = combine_evaluation_standards(evaluation_method_path,invalid_path,2)
|
||||
print(json.dumps(res, ensure_ascii=False, indent=4))
|
||||
end_time=time.time()
|
||||
print("elapsed time:"+str(end_time-start_time))
|
||||
@ -68,35 +68,75 @@ post_process 函数尝试将长字符串按特定模式分割成块，每块至
|
||||
"""
|
||||
|
||||
# 读取JSON数据，提取内容，转换结构，并打印结果
|
||||
def extract_from_notice(merged_baseinfo_path,clause_path, type):
|
||||
if type == 1:
|
||||
target_values = ["投标","投标文件","响应文件"]
|
||||
elif type == 2:
|
||||
# target_values = ["开标", "评标", "定标","磋商程序","中标"]
|
||||
target_values=["开标", "评标", "定标","评审","成交","合同","磋商程序", "中标", "程序", "步骤"]
|
||||
elif type == 3:
|
||||
target_values = ["重新招标、不再招标和终止招标","重新招标","重新采购", "不再招标", "不再采购","终止招标","终止采购"]
|
||||
elif type == 4:
|
||||
target_values = ["评标"] # 测试
|
||||
else:
|
||||
raise ValueError(
|
||||
"Invalid type specified. Use 1 for '投标文件, 投标' or 2 for '开标, 评标, 定标'or 3 for '重新招标'")
|
||||
with open(clause_path, 'r', encoding='utf-8') as file:
|
||||
data = json.load(file)
|
||||
extracted_data = extract_between_sections(data, target_values) #先使用大章节'二、投标文件'这种筛选
|
||||
if not extracted_data:
|
||||
extracted_data = extract_json(data, target_values) # 若没有，再使用'3.投标文件' 筛选
|
||||
def extract_from_notice(merged_baseinfo_path, clause_path, type):
|
||||
"""
|
||||
从公告中提取特定类型的内容。
|
||||
|
||||
Args:
|
||||
merged_baseinfo_path (str): 合并后的基础信息路径。
|
||||
clause_path (str): 包含条款的JSON文件路径。
|
||||
type (int): 提取的类型。
|
||||
1 - ["投标", "投标文件", "响应文件"]
|
||||
2 - ["开标", "评标", "定标", "评审", "成交", "合同", "磋商程序", "中标", "程序", "步骤"]
|
||||
3 - ["重新招标、不再招标和终止招标", "重新招标", "重新采购", "不再招标", "不再采购", "终止招标", "终止采购"]
|
||||
4 - ["评标"] # 测试
|
||||
|
||||
Returns:
|
||||
dict 或 str: 提取并处理后的数据，或在 `clause_path` 为空或发生错误时返回空字符串 `""`。
|
||||
"""
|
||||
# 定义默认的返回结果
|
||||
DEFAULT_RESULT = ""
|
||||
|
||||
# 映射 type 到 target_values
|
||||
type_target_map = {
|
||||
1: ["投标", "投标文件", "响应文件"],
|
||||
2: ["开标", "评标", "定标", "评审", "成交", "合同", "磋商程序", "中标", "程序", "步骤"],
|
||||
3: ["重新招标、不再招标和终止招标", "重新招标", "重新采购", "不再招标", "不再采购", "终止招标", "终止采购"],
|
||||
4: ["评标"] # 测试
|
||||
}
|
||||
|
||||
# 获取对应 type 的 target_values
|
||||
target_values = type_target_map.get(type)
|
||||
if not target_values:
|
||||
print(f"Error: Invalid type specified: {type}. Use 1, 2, 3, or 4.")
|
||||
return DEFAULT_RESULT
|
||||
|
||||
try:
|
||||
# 检查 clause_path 是否为空或仅包含空白字符
|
||||
if clause_path and clause_path.strip():
|
||||
with open(clause_path, 'r', encoding='utf-8') as file:
|
||||
data = json.load(file)
|
||||
|
||||
# 先尝试使用大章节筛选
|
||||
extracted_data = extract_between_sections(data, target_values)
|
||||
|
||||
if not extracted_data:
|
||||
final_result = get_requirements_with_gpt(merged_baseinfo_path, type) # 万一都没，那就调用大模型
|
||||
return final_result
|
||||
final_result=extract_sections(extracted_data,target_values) #后处理，生成键名
|
||||
return final_result
|
||||
# 如果大章节筛选失败，尝试使用另一种筛选方法
|
||||
extracted_data = extract_json(data, target_values)
|
||||
|
||||
if not extracted_data:
|
||||
# 如果所有筛选方法均失败，调用回退函数
|
||||
final_result = get_requirements_with_gpt(merged_baseinfo_path, type)
|
||||
return final_result
|
||||
else:
|
||||
# 后处理，生成键名
|
||||
final_result = extract_sections(extracted_data, target_values)
|
||||
return final_result
|
||||
else:
|
||||
# 合并键值对，启用结构化
|
||||
extracted_data_concatenated = {
|
||||
section: concatenate_keys_values(content)
|
||||
for section, content in extracted_data.items()
|
||||
}
|
||||
return extracted_data_concatenated
|
||||
else:
|
||||
extracted_data_concatenated = {
|
||||
section: concatenate_keys_values(content)
|
||||
for section, content in extracted_data.items()
|
||||
}
|
||||
return extracted_data_concatenated
|
||||
# 如果 clause_path 为空，直接调用回退函数
|
||||
final_result = get_requirements_with_gpt(merged_baseinfo_path, type)
|
||||
return final_result
|
||||
|
||||
except Exception as e:
|
||||
print(f"Error occurred while processing clause_path '{clause_path}': {e}")
|
||||
return DEFAULT_RESULT
|
||||
# print(json.dumps(res, ensure_ascii=False, indent=4))
|
||||
# sorted_data = sort_clean_data_keys(extracted_data) # 对输入的字典 data 的键进行预处理和排序
|
||||
# transformed_data = transform_json(sorted_data)
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user