ly/langchain_projectagent
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

更新最新代码

Browse Source
This commit is contained in:
paituo
2025-07-04 13:52:08 +08:00
parent af64d962eb
commit be98bb1495
52 changed files with 139981 additions and 5788 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/Ontology_Layer.txt
+361
View File
@@ -0,0 +1,361 @@
1.实体类型
EngineeringData
name: STRING
ProjectDivisionSet
name: STRING
ProjectDivisionTree
name: STRING
original_second_level: STRING
original_first_level: STRING
ProjectDivisionItem
type: STRING
取费表id: STRING
name: STRING
序号: STRING
取费表: STRING
GUID: STRING
专业类型: STRING
费率: STRING
资源库名称: STRING
代码: STRING
notCheck: STRING
颜色标记: STRING
编码: STRING
最小资源库编码: STRING
原合价: STRING
备注: STRING
List
合价不含税: STRING
数量: STRING
type: STRING
资源库名称: STRING
工作内容: STRING
取费表类型: STRING
类型: STRING
取费表名称: STRING
单位: STRING
清单名称: STRING
编码: STRING
计算规则: STRING
清单全码: STRING
GUID: STRING
guid: STRING
单价不含税: STRING
单价: STRING
name: STRING
项目特征: STRING
计算式: STRING
合价: STRING
取费表: STRING
ProjectQuantity
数量: STRING
资源库名称: STRING
合价不含税: STRING
特征段: STRING
费用类型: STRING
基准价含税: STRING
结算市场价不含税: STRING
投标单价: STRING
单位: STRING
结算市场价含税: STRING
基准价不含税: STRING
颜色标记: STRING
投标数量: STRING
编码: STRING
类型: STRING
规格型号: STRING
关联父级量: STRING
name: STRING
单价不含税: STRING
损耗率: STRING
截面积: STRING
线重: STRING
id: STRING
供货方: STRING
单重: STRING
集中配送: STRING
市场价含税: STRING
制造长度: STRING
单价含税: STRING
市场价不含税: STRING
增值税率: STRING
合价含税: STRING
运杂费率: STRING
设备类型: STRING
计算式: STRING
人工系数: STRING
定额范围: STRING
定额系数: STRING
基价: STRING
机械费: STRING
人工费: STRING
材料系数: STRING
中标计算式: STRING
机械系数: STRING
材料费: STRING
投标合价: STRING
所属定额库: STRING
批注: STRING
标记: STRING
定额调整系数: STRING
监造物料: STRING
备注: STRING
Quota
特征段: STRING
人工系数: STRING
数量: STRING
定额范围: STRING
定额系数: STRING
合价不含税: STRING
基价: STRING
机械费: STRING
资源库名称: STRING
人工费: STRING
材料系数: STRING
投标单价: STRING
中标计算式: STRING
费用类型: STRING
机械系数: STRING
类型: STRING
材料费: STRING
单位: STRING
颜色标记: STRING
投标合价: STRING
编码: STRING
关联父级量: STRING
投标数量: STRING
计算式: STRING
id: STRING
name: STRING
单价不含税: STRING
所属定额库: STRING
批注: STRING
标记: STRING
定额调整系数: STRING
颜色标记: STRING
备注: STRING
MainMaterial
数量: STRING
资源库名称: STRING
合价不含税: STRING
特征段: STRING
费用类型: STRING
基准价含税: STRING
结算市场价不含税: STRING
投标单价: STRING
单位: STRING
结算市场价含税: STRING
基准价不含税: STRING
颜色标记: STRING
投标数量: STRING
编码: STRING
类型: STRING
规格型号: STRING
关联父级量: STRING
name: STRING
单价不含税: STRING
损耗率: STRING
截面积: STRING
线重: STRING
id: STRING
供货方: STRING
单重: STRING
集中配送: STRING
市场价含税: STRING
制造长度: STRING
单价含税: STRING
市场价不含税: STRING
增值税率: STRING
合价含税: STRING
Equipment
特征段: STRING
单价含税: STRING
单位: STRING
资源库名称: STRING
合价不含税: STRING
类型: STRING
投标数量: STRING
投标单价: STRING
关联父级量: STRING
颜色标记: STRING
运杂费率: STRING
设备类型: STRING
编码: STRING
供货方: STRING
规格型号: STRING
单价不含税: STRING
id: STRING
name: STRING
数量: STRING
计算式: STRING
合价含税: STRING
MaterialOrEquipment
type: STRING
预算价不含税: STRING
单位: STRING
结算市场价不含税: STRING
暂估价: STRING
编码: STRING
结算市场价含税: STRING
全口径市场价不含税: STRING
全口径市场价含税: STRING
是否未计价: STRING
unique_id: STRING
供货方: STRING
结算预算价含税: STRING
结算预算价不含税: STRING
市场价含税: STRING
预算价含税: STRING
id: STRING
name: STRING
数量: STRING
市场价不含税: STRING
拆分: STRING
商品砼: STRING
children: STRING
CostSet
name: STRING
GUID: STRING
CostItem
name: STRING
cost: STRING
unique_id: STRING
id: STRING
FeeTableTemplateSet
name: STRING
typeList: STRING
FeeTableTemplateItem
type: STRING
name: STRING
profession: STRING
outlayID: STRING
FeeCollection
name: STRING
serialNumber: STRING
base: STRING
code: STRING
rate: STRING
remark: STRING
FeeScheduleSet
name: STRING
FeeScheduleItem
name: STRING
ProjectPropertySet
name: STRING
ProjectProperty
特殊地区: STRING
调差选择所在地: STRING
工程所在地: STRING
编制时间: STRING
工程版本: STRING
项目划分: STRING
工程阶段: STRING
工程名称: STRING
专业类型: STRING
地区类型: STRING
组价方式: STRING
人工调差系数: STRING
调差选择地区类型: STRING
机械调差系数: STRING
架线类型: STRING
安装机械调差系数: STRING
清单规范: STRING
市场价唯一: STRING
材料调差系数: STRING
甲供材料计入综合单价: STRING
人工按系数调差: STRING
最高投标限价(万元): STRING
工程总投资: STRING
是否按单位控制工程量精度: STRING
安装人工调差系数: STRING
不同土质定额归属不同清单: STRING
住房公积金缴费费率: STRING
是否是合并工程: STRING
招标人: STRING
相同清单合并: STRING
执行规范: STRING
安装材料调差系数: STRING
拆除调差系数年份: STRING
调差系数年份: STRING
工程税率: STRING
甲供材料计入本体: STRING
软件名称: STRING
电压等级: STRING
预算类型: STRING
本期台数: STRING
建筑人工调差系数: STRING
单台容量: STRING
建筑拆除人工调差系数: STRING
配置选项: STRING
施工企业配合调试费费率: STRING
安装其他设备运杂费率: STRING
安装材机调差系数: STRING
工程性质: STRING
表头设置: STRING
工程静态投资(万元): STRING
安装主要设备运杂费率: STRING
编制依据: STRING
工程动态投资(万元): STRING
基本预备费费率: STRING
阶段类型: STRING
BCL版本: STRING
社会保险费缴费费率: STRING
2. 实体间的关系
(:EngineeringData)-[:HAS_CHILD]->(:ProjectPropertySet)
(:EngineeringData)-[:HAS_CHILD]->(:FeeScheduleSet)
(:EngineeringData)-[:HAS_CHILD]->(:FeeTableTemplateSet)
(:EngineeringData)-[:HAS_CHILD]->(:ProjectDivisionSet)
(:ProjectDivisionSet)-[:HAS_CHILD]->(:ProjectDivisionSet)
(:ProjectDivisionSet)-[:HAS_CHILD]->(:ProjectDivisionItem)
(:ProjectDivisionItem)-[:HAS_CHILD]->(:List)
(:ProjectDivisionItem)-[:HAS_CHILD]->(:ProjectDivisionItem)
(:ProjectDivisionItem)-[:USE]->(:CostSet)
(:ProjectDivisionItem)-[:HAS_CHILD]->(:ProjectQuantity)
(:ProjectDivisionItem)-[:HAS_CHILD]->(:Quota)
(:ProjectDivisionItem)-[:HAS_CHILD]->(:MainMaterial)
(:ProjectDivisionItem)-[:HAS_CHILD]->(:Equipment)
(:ProjectQuantity)-[:HAS_CHILD]->(:MaterialOrEquipment)
(:FeeTableTemplateSet)-[:HAS_CHILD]->(:FeeTableTemplateSet)
(:FeeTableTemplateSet)-[:HAS_CHILD]->(:FeeTableTemplateItem)
(:FeeTableTemplateItem)-[:HAS_CHILD]->(:FeeCollection)
(:FeeCollection)-[:HAS_CHILD]->(:FeeCollection)
(:FeeScheduleSet)-[:HAS_CHILD]->(:FeeScheduleItem)
(:FeeScheduleItem)-[:HAS_CHILD]->(:Fee)
(:Fee)-[:HAS_CHILD]->(:Fee)
(:ProjectPropertySet)-[:HAS_CHILD]->(:ProjectProperty)
(:List)-[:HAS_CHILD]->(:ProjectQuantity)
(:List)-[:HAS_CHILD]->(:Equipment)
(:List)-[:HAS_CHILD]->(:MainMaterial)
(:List)-[:HAS_CHILD]->(:Quota)
(:List)-[:USE]->(:CostSet)
(:Quota)-[:HAS_CHILD]->(:MaterialOrEquipment)
(:CostSet)-[:HAS_CHILD]->(:CostItem)
(:ProjectDivisionTree)-[:HAS_CHILD]->(:ProjectDivisionItem)
(:ProjectQuantity)-[:HAS_CHILD]->(:ProjectQuantity)
(:ProjectQuantity)-[:HAS_CHILD]->(:MainMaterial)
(:MainMaterial)-[:HAS_CHILD]->(:ProjectQuantity)
(:MainMaterial)-[:HAS_CHILD]->(:MainMaterial)
(:ProjectQuantity)-[:USE]->(:CostSet)
(:ProjectDivisionTree)-[:USE]->(:CostSet)
(:ProjectDivisionSet)-[:HAS_CHILD]->(:ProjectDivisionTree)
tools/build_kg_ontolo.py
+1292
View File
File diff suppressed because it is too large Load Diff
tools/streamlt_showcode.py
+167
View File
@@ -0,0 +1,167 @@
import streamlit as st
import os
import sys
import json
import logging
import os
from datetime import datetime
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
# 获取当前时间，格式化为字符串
now_str = datetime.now().strftime("%Y%m%d%H%M%S")
current_file = os.path.splitext(os.path.basename(__file__))[0]
log_filename = f"{current_file}_{now_str}.log"
logging.basicConfig(
level=logging.DEBUG,
format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
handlers=[
logging.FileHandler(os.path.join("logs", log_filename), encoding="utf-8"),
logging.StreamHandler()
],
)
logger = logging.getLogger(current_file)
def setup_logger(logger_name):
"""
设置指定名称的logger，将其级别设置为WARNING并禁用传播
:param logger_name: logger的名称
"""
logger = logging.getLogger(logger_name)
logger.setLevel(logging.WARNING) # 设置httpcore及其子模块的级别
logger.propagate = False # 可选：禁用传播（防止被根logger处理）
return logger
logger_names = ["httpx", "openai", "langsmith.client", "neo4j", "urllib3", "httpcore"]
for name in logger_names:
setup_logger(name)
from src.config import Config
from src.document_loader import load_file
from src.embedding_client import EmbeddingClient
from src.multi_llm_client import MultiAPIKeyChatOpenAI
from src.code_executor import CodeExecutor
from src.neo4j_raw_retriever import Neo4jRawRetriever
from src.prompt_manager import PromptManager
from src.project import ProjectBuilder, ProjectToolkit
from src.project_implementation import ProjectToolkitNeo4j
from src.code_executor import CodeExecutor
config = Config()
business_structure = load_file(config.business_object_structure_path)
bowei_api_docs = load_file(config.bowei_api_docs_path)
#llm_client = MultiAPIKeyChatOpenAI(config.openai)
llm_client_coder = MultiAPIKeyChatOpenAI(config.openai_coder)
prompt_manager = PromptManager()
neo4j_conf = config.neo4j_conf
embedding_conf = config.embedding
embedding_client = EmbeddingClient(embedding_conf)
# 创建Neo4j检索器
knowledge_retriever = Neo4jRawRetriever(neo4j_conf)
ProjectBuilder.register(ProjectToolkitNeo4j, knowledge_retriever.driver)
code_executor = CodeExecutor(prompt_manager.prompts, llm_client_coder, config.max_retries)
def load_jsonl(file_path):
"""加载JSONL文件并返回JSON记录列表"""
records = []
try:
with open(file_path, 'r', encoding='utf-8') as file:
for line in file:
if line.strip():
records.append(json.loads(line))
return records
except Exception as e:
st.error(f"加载文件失败: {str(e)}")
return []
def run_code(data):
global code_executor
"""运行JSON记录中的代码并返回结果"""
if not data or 'code' not in data:
return {
"code": 40000,
"message": "没有可执行的代码",
"status": False,
"data": None
}
result = code_executor.execute_code(data['code'])
def main():
st.set_page_config(layout="wide", page_title="JSONL查看器")
st.title("JSONL文件查看器")
# 设置默认文件路径
default_file_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), "../tests", "code.jsonl")
# 文件路径输入
file_path = st.text_input("JSONL文件路径", value=default_file_path)
if not os.path.exists(file_path):
st.warning(f"文件不存在: {file_path}")
return
# 加载JSONL文件
records = load_jsonl(file_path)
if not records:
st.warning("没有找到有效的记录")
return
# 创建两列布局
col1, col2 = st.columns([1, 3])
# 左侧列表
with col1:
st.subheader("记录列表")
selected_index = None
for i, record in enumerate(records):
if st.button(record.get('name', f"记录 {i+1}"), key=f"btn_{i}"):
selected_index = i
# 右侧详细信息
with col2:
if 'selected_index' not in st.session_state:
st.session_state.selected_index = 0
if selected_index is not None:
st.session_state.selected_index = selected_index
if st.session_state.selected_index < len(records):
selected_record = records[st.session_state.selected_index]
st.subheader(f"查询问题: {selected_record.get('name', '无名称')}")
st.info(selected_record.get('query', '无查询信息'))
st.subheader("代码")
st.code(selected_record.get('code', '无代码'), language='python')
# 运行代码按钮
if st.button("运行代码"):
with st.spinner('正在执行代码...'):
result = run_code(selected_record)
st.subheader("运行结果")
if result.get('status'):
st.success("执行成功")
st.json(result.get('data'))
else:
st.error(f"执行失败: {result.get('message', '未知错误')}")
if __name__ == "__main__":
main()
tools/supplement_kg.py
+117
View File
@@ -0,0 +1,117 @@
import json
from typing import Dict, List, Any
import copy
class ExpenseProcessor:
def __init__(self):
pass
@staticmethod
def calculate_parent_costs(node: Dict[str, Any]) -> List[Dict[str, Any]]:
if "children" not in node:
if "id" in node and "cost" in node:
return [{"id": node["id"], "cost": node["cost"]}]
elif "cost" in node:
return [{"cost": node["cost"]}]
return []
result_nodes = []
processed_ids = set()
for child in node["children"]:
child_costs = ExpenseProcessor.calculate_parent_costs(child)
for cost_item in child_costs:
if "id" in cost_item:
found = False
for existing in result_nodes:
if "id" in existing and existing["id"] == cost_item["id"]:
existing["cost"] = str(float(existing["cost"]) + float(cost_item["cost"]))
found = True
break
if not found:
result_nodes.append(copy.deepcopy(cost_item))
processed_ids.add(cost_item["id"])
else:
found = False
for existing in result_nodes:
if "id" not in existing:
existing["cost"] = str(float(existing["cost"]) + float(cost_item["cost"]))
found = True
break
if not found:
result_nodes.append(copy.deepcopy(cost_item))
return result_nodes
@staticmethod
def process_node(node: Dict[str, Any]) -> Dict[str, Any]:
result = copy.deepcopy(node)
if "children" not in node or not node["children"]:
return result
cost_items = ExpenseProcessor.calculate_parent_costs(node)
if cost_items:
result["sum"] = cost_items
result["children"] = [ExpenseProcessor.process_node(child) for child in node["children"]]
return result
@staticmethod
def process_expense_preview(expense_preview: Dict[str, Any]) -> Dict[str, Any]:
result = copy.deepcopy(expense_preview)
for category_key, category_value in expense_preview.items():
for subcategory_key, subcategory_value in category_value.items():
if isinstance(subcategory_value, list):
result[category_key][subcategory_key] = [
ExpenseProcessor.process_node(item) for item in subcategory_value
]
return result
@classmethod
def load_and_process_from_file(cls, input_path: str, output_path: str = None) -> Dict[str, Any]:
"""
从文件加载 JSON 并处理
:param input_path: 输入文件路径
:param output_path: 输出文件路径（可选）
:return: 处理后的完整数据
"""
with open(input_path, "r", encoding="utf-8") as f:
data = json.load(f)
if "projectData" in data and "expensePreview" in data["projectData"]:
processed_data = copy.deepcopy(data)
processed_data["projectData"]["expensePreview"] = cls.process_expense_preview(
data["projectData"]["expensePreview"]
)
if output_path:
with open(output_path, "w", encoding="utf-8") as f:
json.dump(processed_data, f, ensure_ascii=False, indent=4)
print(f"处理完成，结果已保存到 {output_path}")
return processed_data
else:
raise ValueError("未找到 projectData.expensePreview 路径")
@classmethod
def process_raw_data(cls, raw_data: Dict[str, Any]) -> Dict[str, Any]:
"""
直接处理原始数据（不涉及文件读写）
:param raw_data: 原始数据，格式应包含 projectData.expensePreview
:return: 处理后的数据
"""
if "projectData" in raw_data and "expensePreview" in raw_data["projectData"]:
processed_data = copy.deepcopy(raw_data)
processed_data["projectData"]["expensePreview"] = cls.process_expense_preview(
raw_data["projectData"]["expensePreview"]
)
return processed_data
else:
raise ValueError("未找到 projectData.expensePreview 路径")
if __name__ == "__main__":
input_file = "dataset/json/主网预算/架空_clean.json" # 输入 JSON 文件路径
output_file = "dataset/json/主网预算/output.json" # 输出 JSON 文件路径
# 使用类方法加载并处理 JSON 文件
ExpenseProcessor.load_and_process_from_file(input_file, output_file)
tools/xml_to_json.py
+206
View File
@@ -0,0 +1,206 @@
import chardet
import xml.etree.ElementTree as ET
import json
import re
def clean_bracketed_strings(input_str: str) -> str:
# 替换【'xxx'】为【xxx】
result = re.sub(r"【'([^']+)'】", r"【\1】", input_str)
# 替换【['xxx']】为【xxx】
result = re.sub(r"【\['([^']+)'\]】", r"【\1】", result)
# 替换【['xxx','yyy']】为【xxx,yyy】（多个项的情况）
result = re.sub(r"【\[((?:'[^']+',?)+)\]】", lambda m: "【" + m.group(1).replace("'", "") + "】", result)
return result
def read_xml_as_string(file_path):
# 先读取部分字节探测编码
with open(file_path, 'rb') as f:
raw_data = f.read()
result = chardet.detect(raw_data)
encoding = result['encoding']
# 使用探测到的编码重新读取为字符串
return raw_data.decode(encoding)
def parse_keyword(keyword, indicator_name):
# 特殊处理：电压等级
# if indicator_name == "电压等级":
# return {"映射规则": "1", "指标映射": [keyword]}
# 处理范围表达式（包含"||"分隔符）
if "||" in keyword:
parts = keyword.split("||")
table_rows = []
all_codes = []
for part in parts:
if "@@" not in part:
continue
codes_str, value = part.split("@@", 1)
code_ranges = codes_str.split("、")
for code_range in code_ranges:
# 处理连续编号（如YX5-67~69）
if "~" in code_range:
prefix, range_part = code_range.rsplit("-", 1)
start_str, end_str = range_part.split("~")
try:
start = int(start_str)
end = int(end_str)
for num in range(start, end + 1):
all_codes.append(f"{prefix}-{num}")
except ValueError:
all_codes.append(code_range)
else:
all_codes.append(code_range)
table_rows.append(f"| {codes_str} | {value} |")
rule_table = "| 资源识别规则 | 指标值 |\n|-------|-------|\n" + "\n".join(table_rows)
return {"映射规则": rule_table, "指标映射": all_codes}
# 处理数学公式（包含"/"和括号）
if "/" in keyword and "(" in keyword and ")" in keyword:
# 提取分子（括号前部分）
molecule = keyword.split("/")[0].strip()
# 提取分母（括号内部分）
denominator_start = keyword.find("(") + 1
denominator_end = keyword.find(")")
denominator_expr = keyword[denominator_start:denominator_end]
# 分割分母中的加法项
denominator_items = [item.strip() for item in denominator_expr.split("+")]
return {"映射规则": keyword, "指标映射": [molecule] + denominator_items}
# 处理加法表达式
if "+" in keyword:
items = [item.strip() for item in keyword.split("+")]
return {"映射规则": keyword, "指标映射": items}
# 默认处理（普通关键字）
return {"映射规则": keyword, "指标映射": [keyword]}
def xml_to_json(xml_content, output_path):
root = ET.fromstring(xml_content)
records = root.findall('.//records/record')
result = []
# 定义需要特殊处理的数据来源类型
scope_based_sources = ["主材单价", "主材参数", "主材数量", "定额参数", "定额数量", "工程费用"]
direct_sources = ["报表指标", "指标库"]
project_division = ["项目划分费用"]
for record in records:
unit = record.get("单位")
unit_type = record.get("单价类型")
order = record.get("序号")
extraction_method = record.get("提取方式")
indicator_type = record.get("指标类型")
index_extraction_scope = record.get("指标提取范围")
data_sources = record.get("数据来源")
indicator_name = record.get("指标名称")
keyword = record.get("关键字")
parsed = parse_keyword(keyword, indicator_name)
base_item = {
"指标名称": indicator_name,
"code": "",
"单位": unit,
"单价类型": unit_type,
"序号": order,
"提取方式": extraction_method,
"指标类型": indicator_type,
"数据来源": data_sources
}
if data_sources in direct_sources:
base_item["指标描述"] = {
"指标映射": parsed["指标映射"],
"映射规则": parsed["映射规则"]
}
result.append(base_item)
elif data_sources in project_division:
mapping_desc = f"从【{index_extraction_scope}】项目划分中获取名称属于【{indicator_name}】的费用"
base_item["指标描述"] = {
"指标映射": mapping_desc,
"映射规则": parsed["映射规则"]
}
result.append(base_item)
elif data_sources in scope_based_sources:
temp0 = parsed["指标映射"]
temp1 = parsed["映射规则"]
if index_extraction_scope:
# 取数据来源的最后两个字（如"单价"、"参数"等）
temp2 = data_sources[-2:]
# 取数据来源的开头两个字（如"定额"、"清单"等）
temp3 = data_sources[0:2]
if temp3 in ["清单", "定额", "人材机"]:
mapping_desc = f"从【{index_extraction_scope}】及其子孙项目划分中查找编码中包含【{temp0}】的所有【{temp3}】的【{temp2}】之和"
elif temp3 in ["主材", "设备"]:
mapping_desc = f"从【{index_extraction_scope}】及其子孙项目划分中查找名称中包含【{temp0}】的所有【{temp3}】的【{temp2}】之和"
else:
mapping_desc = f"从【{data_sources}】中获取{temp0}的属性"
base_item["指标描述"] = {
"指标映射": mapping_desc,
"映射规则": temp1
}
result.append(base_item)
else:
# 处理未定义的数据来源类型
base_item["指标描述"] = {
"指标映射": parsed["指标映射"],
"映射规则": parsed["映射规则"]
}
result.append(base_item)
for item in result:
desc = item.get("指标描述", {})
mapping_rule = desc.get("映射规则")
indicator_map = desc.get("指标映射")
if isinstance(mapping_rule, str) and isinstance(indicator_map, str):
match = re.search(r"@([^\.]+)\.", mapping_rule)
if match:
source = match.group(1)
# 替换【工程费用】为提取的 source（如【其他费用】）
new_mapping = re.sub(r"【.*?】", f"【{source}】", indicator_map)
new_mapping = re.sub(r"@[^.]+\.", "", new_mapping)
item["指标描述"]["指标映射"] = new_mapping.replace('[', '【').replace(']', '】')
if isinstance(indicator_map, list):
first_elem = indicator_map[0]
if first_elem.startswith("(") and first_elem.endswith(")"):
del indicator_map[0]
# 保存为 JSON 文件
with open(output_path, 'w', encoding='utf-8') as f:
json.dump(result, f, ensure_ascii=False, indent=2)
for item in result:
desc = item.get("指标描述", {})
mapping_rule = desc.get("映射规则")
indicator_map = desc.get("指标映射")
if isinstance(indicator_map, str):
new_mapping = clean_bracketed_strings(indicator_map)
item["指标描述"]["指标映射"] = new_mapping
# 保存为 JSON 文件
with open(output_path, 'w', encoding='utf-8') as f:
json.dump(result, f, ensure_ascii=False, indent=2)
return "结果已保存"
xml_content = read_xml_as_string('dataset/主网架空线路造价分析指标.xml')
json_output = xml_to_json(xml_content, output_path= "./tests/zhibiao.json")
print("转换完毕！")