Remove namespace_prefix from PostgreSQL, maintain consistency with other storage implementation

env.example

@@ -5,7 +5,6 @@
 # PORT=9621
 # WORKERS=2
 ### separating data from difference Lightrag instances
-# NAMESPACE_PREFIX=lightrag
 ### Max nodes return from grap retrieval
 # MAX_GRAPH_NODES=1000
 # CORS_ORIGINS=http://localhost:3000,http://localhost:8080

lightrag/api/lightrag_server.py

@@ -315,7 +315,7 @@ def create_app(args):
                 "similarity_threshold": 0.95,
                 "use_llm_check": False,
             },
-            namespace_prefix=args.namespace_prefix,
+            # namespace_prefix=args.namespace_prefix,
             auto_manage_storages_states=False,
             max_parallel_insert=args.max_parallel_insert,
         )
@@ -345,7 +345,7 @@ def create_app(args):
                 "similarity_threshold": 0.95,
                 "use_llm_check": False,
             },
-            namespace_prefix=args.namespace_prefix,
+            # namespace_prefix=args.namespace_prefix,
             auto_manage_storages_states=False,
             max_parallel_insert=args.max_parallel_insert,
         )

lightrag/kg/postgres_impl.py

@@ -254,8 +254,6 @@ class PGKVStorage(BaseKVStorage):
     db: PostgreSQLDB = field(default=None)
 
     def __post_init__(self):
-        namespace_prefix = self.global_config.get("namespace_prefix")
-        self.base_namespace = self.namespace.replace(namespace_prefix, "")
         self._max_batch_size = self.global_config["embedding_batch_num"]
 
     async def initialize(self):
@@ -271,7 +269,7 @@ class PGKVStorage(BaseKVStorage):
 
     async def get_by_id(self, id: str) -> dict[str, Any] | None:
         """Get doc_full data by id."""
-        sql = SQL_TEMPLATES["get_by_id_" + self.base_namespace]
+        sql = SQL_TEMPLATES["get_by_id_" + self.namespace]
         params = {"workspace": self.db.workspace, "id": id}
         if is_namespace(self.namespace, NameSpace.KV_STORE_LLM_RESPONSE_CACHE):
             array_res = await self.db.query(sql, params, multirows=True)
@@ -285,7 +283,7 @@ class PGKVStorage(BaseKVStorage):
 
     async def get_by_mode_and_id(self, mode: str, id: str) -> Union[dict, None]:
         """Specifically for llm_response_cache."""
-        sql = SQL_TEMPLATES["get_by_mode_id_" + self.base_namespace]
+        sql = SQL_TEMPLATES["get_by_mode_id_" + self.namespace]
         params = {"workspace": self.db.workspace, mode: mode, "id": id}
         if is_namespace(self.namespace, NameSpace.KV_STORE_LLM_RESPONSE_CACHE):
             array_res = await self.db.query(sql, params, multirows=True)
@@ -299,7 +297,7 @@ class PGKVStorage(BaseKVStorage):
     # Query by id
     async def get_by_ids(self, ids: list[str]) -> list[dict[str, Any]]:
         """Get doc_chunks data by id"""
-        sql = SQL_TEMPLATES["get_by_ids_" + self.base_namespace].format(
+        sql = SQL_TEMPLATES["get_by_ids_" + self.namespace].format(
             ids=",".join([f"'{id}'" for id in ids])
         )
         params = {"workspace": self.db.workspace}
@@ -320,7 +318,7 @@ class PGKVStorage(BaseKVStorage):
 
     async def get_by_status(self, status: str) -> Union[list[dict[str, Any]], None]:
         """Specifically for llm_response_cache."""
-        SQL = SQL_TEMPLATES["get_by_status_" + self.base_namespace]
+        SQL = SQL_TEMPLATES["get_by_status_" + self.namespace]
         params = {"workspace": self.db.workspace, "status": status}
         return await self.db.query(SQL, params, multirows=True)
 
@@ -403,8 +401,6 @@ class PGVectorStorage(BaseVectorStorage):
 
     def __post_init__(self):
         self._max_batch_size = self.global_config["embedding_batch_num"]
-        namespace_prefix = self.global_config.get("namespace_prefix")
-        self.base_namespace = self.namespace.replace(namespace_prefix, "")
         config = self.global_config.get("vector_db_storage_cls_kwargs", {})
         cosine_threshold = config.get("cosine_better_than_threshold")
         if cosine_threshold is None:
@@ -533,7 +529,7 @@ class PGVectorStorage(BaseVectorStorage):
         else:
             formatted_ids = "NULL"
 
-        sql = SQL_TEMPLATES[self.base_namespace].format(
+        sql = SQL_TEMPLATES[self.namespace].format(
             embedding_string=embedding_string, doc_ids=formatted_ids
         )
         params = {

lightrag/lightrag.py

@@ -229,6 +229,7 @@ class LightRAG:
     vector_db_storage_cls_kwargs: dict[str, Any] = field(default_factory=dict)
     """Additional parameters for vector database storage."""
 
+    # TODO：deprecated, remove in the future, use WORKSPACE instead
     namespace_prefix: str = field(default="")
     """Prefix for namespacing stored data across different environments."""