上传文件

nodetype_ex.py

@@ -0,0 +1,168 @@
+from neo4j import GraphDatabase
+import pandas as pd
+import json
+
+
+class KnowledgeGraphQuerier:
+    def __init__(self, uri, user, password):
+        self.driver = GraphDatabase.driver(uri, auth=(user, password))
+
+    def close(self):
+        self.driver.close()
+
+    def get_node_name(self, node_types, properties):
+        """根据节点类型和可用属性确定节点名称"""
+        # 根据用户提供的信息，为不同类型的节点设置名称属性
+        priority_map = {
+            "ProjectDivisionSet": ["name"],
+            "ProjectDivisionItem": ["name"],
+            "Quota": ["name"],
+            "MainMaterial": ["name"],
+            "Equipment": ["name"],
+            "MaterialOrEquipment": ["name"],
+            "CostSet": ["GUID"],
+            "CostItem": ["id"],
+            "FeeTableTemplateSet": ["name"],
+            "FeeTableTemplateItem": ["name"],
+            "FeeCollection": ["name"],
+            "FeeScheduleSet": ["name"],
+            "FeeScheduleItem": ["name"],
+            "Fee": ["name"],
+            "ProjectPropertySet": ["name"],
+            "ProjectProperty": ["软件名称"],
+        }
+
+        # 默认优先级顺序（适用于未明确定义的节点类型）
+        default_priority = ["名称", "项目名称", "费用名称", "工程名称", "id", "GUID", "序号", "编码", "代码"]
+
+        # 确定使用哪个优先级列表
+        priority_list = None
+        for node_type in node_types:
+            if node_type in priority_map:
+                priority_list = priority_map[node_type]
+                break
+
+        if not priority_list:
+            priority_list = default_priority
+
+        # 按优先级查找第一个存在的属性
+        for prop in priority_list:
+            if prop in properties and properties[prop]:
+                return properties[prop]
+
+        # 如果没有找到任何匹配的属性，返回第一个非空属性
+        for prop, value in properties.items():
+            if value:
+                return value
+
+        return "无名称"
+
+    def get_simple_hierarchy(self):
+        """获取简洁的图谱层级结构"""
+        # 首先获取所有节点和它们的名称
+        with self.driver.session() as session:
+            result = session.run(
+                """
+                MATCH (n)
+                RETURN id(n) AS node_id, labels(n) AS node_types, properties(n) AS properties
+            """
+            )
+
+            nodes = {}
+            for record in result:
+                node_id = record["node_id"]
+                node_types = record["node_types"]
+                properties = record["properties"]
+
+                # 获取节点名称
+                name = self.get_node_name(node_types, properties)
+
+                # 以第一个类型作为节点的主要类型
+                main_type = node_types[0] if node_types else "Unknown"
+
+                nodes[node_id] = {"id": node_id, "type": main_type, "name": name, "children": []}
+
+        # 获取节点之间的关系
+        with self.driver.session() as session:
+            result = session.run(
+                """
+                MATCH (a)-[r]->(b)
+                RETURN id(a) AS source_id, type(r) AS relationship_type, id(b) AS target_id
+                """
+            )
+
+            # 构建父子关系
+            for record in result:
+                source_id = record["source_id"]
+                target_id = record["target_id"]
+
+                if source_id in nodes and target_id in nodes:
+                    nodes[source_id]["children"].append(target_id)
+
+        # 找出根节点（没有父节点的节点）
+        all_nodes = set(nodes.keys())
+        child_nodes = set()
+        for node in nodes.values():
+            child_nodes.update(node["children"])
+        root_nodes = all_nodes - child_nodes
+
+        # 构建层级结构
+        hierarchy = []
+        for root_id in root_nodes:
+            hierarchy.append(self._build_simple_hierarchy(root_id, nodes))
+
+        # 如果没有根节点，则返回所有节点的平级结构
+        if not hierarchy:
+            for node_id in nodes:
+                node_info = nodes[node_id]
+                hierarchy.append({node_info["type"]: node_info["name"]})
+
+        return hierarchy
+
+    def _build_simple_hierarchy(self, node_id, nodes):
+        """递归构建简洁的节点层级结构"""
+        node = nodes[node_id]
+        node_type = node["type"]
+        node_name = node["name"]
+
+        # 创建当前节点的键值对
+        result = {node_type: node_name}
+
+        # 如果有子节点，递归处理
+        if node["children"]:
+            children = []
+            for child_id in node["children"]:
+                children.append(self._build_simple_hierarchy(child_id, nodes))
+
+            # 将子节点添加到当前节点下
+            result["children"] = children
+
+        return result
+
+
+def main():
+    # 连接信息
+    uri = "bolt://172.20.0.145:7687"
+    user = "neo4j"
+    password = "password"
+
+    querier = KnowledgeGraphQuerier(uri, user, password)
+
+    try:
+        # 获取简洁的图谱层级结构
+        print("正在构建简洁的图谱层级结构...")
+        hierarchy = querier.get_simple_hierarchy()
+
+        # 将层级结构保存为JSON文件
+        print("正在保存图谱层级结构到JSON文件...")
+        with open("kg_simple_hierarchy.json", "w", encoding="utf-8") as f:
+            json.dump(hierarchy, f, ensure_ascii=False, indent=2)
+
+        print("简洁的图谱层级结构已保存到 kg_simple_hierarchy.json")
+
+    finally:
+        querier.close()
+
+
+if __name__ == "__main__":
+    main()