[add] Create community nodes
This commit is contained in:
lanceyq
129
api/app/repositories/neo4j/community_repository.py
Normal file
129
api/app/repositories/neo4j/community_repository.py
Normal file
@@ -0,0 +1,129 @@
|
||||
"""Community 节点仓库
|
||||
|
||||
管理 Neo4j 中 Community 节点及 BELONGS_TO_COMMUNITY 边的 CRUD 操作。
|
||||
"""
|
||||
|
||||
import logging
|
||||
from typing import Dict, List, Optional
|
||||
|
||||
from app.repositories.neo4j.neo4j_connector import Neo4jConnector
|
||||
from app.repositories.neo4j.cypher_queries import (
|
||||
COMMUNITY_NODE_UPSERT,
|
||||
ENTITY_JOIN_COMMUNITY,
|
||||
ENTITY_LEAVE_ALL_COMMUNITIES,
|
||||
GET_ENTITY_NEIGHBORS,
|
||||
GET_ALL_ENTITIES_FOR_USER,
|
||||
GET_COMMUNITY_MEMBERS,
|
||||
CHECK_USER_HAS_COMMUNITIES,
|
||||
UPDATE_COMMUNITY_MEMBER_COUNT,
|
||||
)
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
||||
class CommunityRepository:
|
||||
def __init__(self, connector: Neo4jConnector):
|
||||
self.connector = connector
|
||||
|
||||
async def upsert_community(
|
||||
self, community_id: str, end_user_id: str, member_count: int = 0
|
||||
) -> Optional[str]:
|
||||
"""创建或更新 Community 节点，返回 community_id。"""
|
||||
try:
|
||||
result = await self.connector.execute_query(
|
||||
COMMUNITY_NODE_UPSERT,
|
||||
community_id=community_id,
|
||||
end_user_id=end_user_id,
|
||||
member_count=member_count,
|
||||
)
|
||||
return result[0]["community_id"] if result else None
|
||||
except Exception as e:
|
||||
logger.error(f"upsert_community failed: {e}")
|
||||
return None
|
||||
|
||||
async def assign_entity_to_community(
|
||||
self, entity_id: str, community_id: str, end_user_id: str
|
||||
) -> bool:
|
||||
"""将实体关联到社区（先解除旧关联，再建立新关联）。"""
|
||||
try:
|
||||
await self.connector.execute_query(
|
||||
ENTITY_LEAVE_ALL_COMMUNITIES,
|
||||
entity_id=entity_id,
|
||||
end_user_id=end_user_id,
|
||||
)
|
||||
result = await self.connector.execute_query(
|
||||
ENTITY_JOIN_COMMUNITY,
|
||||
entity_id=entity_id,
|
||||
community_id=community_id,
|
||||
end_user_id=end_user_id,
|
||||
)
|
||||
return bool(result)
|
||||
except Exception as e:
|
||||
logger.error(f"assign_entity_to_community failed: {e}")
|
||||
return False
|
||||
|
||||
async def get_entity_neighbors(
|
||||
self, entity_id: str, end_user_id: str
|
||||
) -> List[Dict]:
|
||||
"""查询实体的直接邻居及其社区归属。"""
|
||||
try:
|
||||
return await self.connector.execute_query(
|
||||
GET_ENTITY_NEIGHBORS,
|
||||
entity_id=entity_id,
|
||||
end_user_id=end_user_id,
|
||||
)
|
||||
except Exception as e:
|
||||
logger.error(f"get_entity_neighbors failed: {e}")
|
||||
return []
|
||||
|
||||
async def get_all_entities(self, end_user_id: str) -> List[Dict]:
|
||||
"""拉取某用户下所有实体及其当前社区归属。"""
|
||||
try:
|
||||
return await self.connector.execute_query(
|
||||
GET_ALL_ENTITIES_FOR_USER,
|
||||
end_user_id=end_user_id,
|
||||
)
|
||||
except Exception as e:
|
||||
logger.error(f"get_all_entities failed: {e}")
|
||||
return []
|
||||
|
||||
async def get_community_members(
|
||||
self, community_id: str, end_user_id: str
|
||||
) -> List[Dict]:
|
||||
"""查询社区成员列表。"""
|
||||
try:
|
||||
return await self.connector.execute_query(
|
||||
GET_COMMUNITY_MEMBERS,
|
||||
community_id=community_id,
|
||||
end_user_id=end_user_id,
|
||||
)
|
||||
except Exception as e:
|
||||
logger.error(f"get_community_members failed: {e}")
|
||||
return []
|
||||
|
||||
async def has_communities(self, end_user_id: str) -> bool:
|
||||
"""检查该用户是否已有 Community 节点（用于判断全量 vs 增量）。"""
|
||||
try:
|
||||
result = await self.connector.execute_query(
|
||||
CHECK_USER_HAS_COMMUNITIES,
|
||||
end_user_id=end_user_id,
|
||||
)
|
||||
return result[0]["community_count"] > 0 if result else False
|
||||
except Exception as e:
|
||||
logger.error(f"has_communities failed: {e}")
|
||||
return False
|
||||
|
||||
async def refresh_member_count(
|
||||
self, community_id: str, end_user_id: str
|
||||
) -> int:
|
||||
"""重新统计并更新社区成员数，返回最新数量。"""
|
||||
try:
|
||||
result = await self.connector.execute_query(
|
||||
UPDATE_COMMUNITY_MEMBER_COUNT,
|
||||
community_id=community_id,
|
||||
end_user_id=end_user_id,
|
||||
)
|
||||
return result[0]["member_count"] if result else 0
|
||||
except Exception as e:
|
||||
logger.error(f"refresh_member_count failed: {e}")
|
||||
return 0
|
||||
@@ -1058,4 +1058,95 @@ Graph_Node_query = """
|
||||
3 AS priority
|
||||
LIMIT $limit
|
||||
|
||||
"""
|
||||
"""
|
||||
|
||||
|
||||
# ============================================================
|
||||
# Community 节点 & BELONGS_TO_COMMUNITY 边
|
||||
# ============================================================
|
||||
|
||||
COMMUNITY_NODE_SAVE = """
|
||||
MERGE (c:Community {community_id: $community_id})
|
||||
SET c.end_user_id = $end_user_id,
|
||||
c.formed_at = $formed_at,
|
||||
c.updated_at = datetime(),
|
||||
c.status = $status,
|
||||
c.member_count = $member_count
|
||||
RETURN c.community_id AS community_id
|
||||
"""
|
||||
|
||||
COMMUNITY_ADD_MEMBER = """
|
||||
MATCH (e:ExtractedEntity {id: $entity_id, end_user_id: $end_user_id})
|
||||
MATCH (c:Community {community_id: $community_id, end_user_id: $end_user_id})
|
||||
MERGE (e)-[:BELONGS_TO_COMMUNITY]->(c)
|
||||
SET c.updated_at = datetime(),
|
||||
c.member_count = $member_count
|
||||
"""
|
||||
|
||||
|
||||
|
||||
# ─── Community 聚类相关 Cypher 模板 ───────────────────────────────────────────
|
||||
|
||||
COMMUNITY_NODE_UPSERT = """
|
||||
MERGE (c:Community {community_id: $community_id})
|
||||
SET c.end_user_id = $end_user_id,
|
||||
c.member_count = $member_count,
|
||||
c.updated_at = datetime()
|
||||
RETURN c.community_id AS community_id
|
||||
"""
|
||||
|
||||
ENTITY_JOIN_COMMUNITY = """
|
||||
MATCH (e:ExtractedEntity {id: $entity_id, end_user_id: $end_user_id})
|
||||
MATCH (c:Community {community_id: $community_id, end_user_id: $end_user_id})
|
||||
MERGE (e)-[:BELONGS_TO_COMMUNITY]->(c)
|
||||
SET c.updated_at = datetime()
|
||||
RETURN e.id AS entity_id, c.community_id AS community_id
|
||||
"""
|
||||
|
||||
ENTITY_LEAVE_ALL_COMMUNITIES = """
|
||||
MATCH (e:ExtractedEntity {id: $entity_id, end_user_id: $end_user_id})
|
||||
MATCH (e)-[r:BELONGS_TO_COMMUNITY]->(:Community)
|
||||
DELETE r
|
||||
"""
|
||||
|
||||
GET_ENTITY_NEIGHBORS = """
|
||||
MATCH (e:ExtractedEntity {id: $entity_id, end_user_id: $end_user_id})
|
||||
OPTIONAL MATCH (e)-[:EXTRACTED_RELATIONSHIP]-(nb:ExtractedEntity {end_user_id: $end_user_id})
|
||||
OPTIONAL MATCH (nb)-[:BELONGS_TO_COMMUNITY]->(c:Community)
|
||||
RETURN DISTINCT
|
||||
nb.id AS id,
|
||||
nb.name AS name,
|
||||
nb.name_embedding AS name_embedding,
|
||||
nb.activation_value AS activation_value,
|
||||
CASE WHEN c IS NOT NULL THEN c.community_id ELSE null END AS community_id
|
||||
"""
|
||||
|
||||
GET_ALL_ENTITIES_FOR_USER = """
|
||||
MATCH (e:ExtractedEntity {end_user_id: $end_user_id})
|
||||
OPTIONAL MATCH (e)-[:BELONGS_TO_COMMUNITY]->(c:Community)
|
||||
RETURN e.id AS id,
|
||||
e.name AS name,
|
||||
e.name_embedding AS name_embedding,
|
||||
e.activation_value AS activation_value,
|
||||
CASE WHEN c IS NOT NULL THEN c.community_id ELSE null END AS community_id
|
||||
"""
|
||||
|
||||
GET_COMMUNITY_MEMBERS = """
|
||||
MATCH (e:ExtractedEntity {end_user_id: $end_user_id})-[:BELONGS_TO_COMMUNITY]->(c:Community {community_id: $community_id})
|
||||
RETURN e.id AS id, e.name AS name, e.entity_type AS entity_type,
|
||||
e.importance_score AS importance_score, e.activation_value AS activation_value,
|
||||
e.name_embedding AS name_embedding
|
||||
ORDER BY coalesce(e.activation_value, 0) DESC
|
||||
"""
|
||||
|
||||
CHECK_USER_HAS_COMMUNITIES = """
|
||||
MATCH (c:Community {end_user_id: $end_user_id})
|
||||
RETURN count(c) AS community_count
|
||||
"""
|
||||
|
||||
UPDATE_COMMUNITY_MEMBER_COUNT = """
|
||||
MATCH (e:ExtractedEntity {end_user_id: $end_user_id})-[:BELONGS_TO_COMMUNITY]->(c:Community {community_id: $community_id})
|
||||
WITH c, count(e) AS cnt
|
||||
SET c.member_count = cnt
|
||||
RETURN c.community_id AS community_id, cnt AS member_count
|
||||
"""
|
||||
|
||||
@@ -1,3 +1,4 @@
|
||||
import asyncio
|
||||
from typing import List
|
||||
|
||||
# 使用新的仓储层
|
||||
@@ -288,6 +289,14 @@ async def save_dialog_and_statements_to_neo4j(
|
||||
}
|
||||
logger.info("Transaction completed. Summary: %s", summary)
|
||||
logger.debug("Full transaction results: %r", results)
|
||||
|
||||
# 写入成功后，触发聚类
|
||||
if entity_nodes:
|
||||
end_user_id = entity_nodes[0].end_user_id
|
||||
new_entity_ids = [e.id for e in entity_nodes]
|
||||
logger.info(f"[Clustering] 准备触发聚类，实体数: {len(new_entity_ids)}, end_user_id: {end_user_id}")
|
||||
await _trigger_clustering(new_entity_ids, end_user_id)
|
||||
|
||||
return True
|
||||
|
||||
except Exception as e:
|
||||
@@ -295,3 +304,28 @@ async def save_dialog_and_statements_to_neo4j(
|
||||
print(f"Neo4j integration error: {e}")
|
||||
print("Continuing without database storage...")
|
||||
return False
|
||||
|
||||
|
||||
async def _trigger_clustering(
|
||||
new_entity_ids: List[str],
|
||||
end_user_id: str,
|
||||
) -> None:
|
||||
"""
|
||||
聚类触发函数，自动判断全量初始化还是增量更新。
|
||||
"""
|
||||
connector = None
|
||||
try:
|
||||
from app.core.memory.storage_services.clustering_engine import LabelPropagationEngine
|
||||
logger.info(f"[Clustering] 开始聚类，end_user_id={end_user_id}, 实体数={len(new_entity_ids)}")
|
||||
connector = Neo4jConnector()
|
||||
engine = LabelPropagationEngine(connector)
|
||||
await engine.run(end_user_id=end_user_id, new_entity_ids=new_entity_ids)
|
||||
logger.info(f"[Clustering] 聚类完成，end_user_id={end_user_id}")
|
||||
except Exception as e:
|
||||
logger.error(f"[Clustering] 聚类触发失败: {e}", exc_info=True)
|
||||
finally:
|
||||
if connector:
|
||||
try:
|
||||
await connector.close()
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
Reference in New Issue
Block a user