fix(GraphRAG): create same graph name error. (#2215)

dbgpt/datasource/conn_tugraph.py

@@ -35,6 +35,19 @@ class TuGraphConnector(BaseConnector):
 
         return not exists
 
+    def is_exist(self, graph_name: str) -> bool:
+        """Check a new graph in the database if it doesn't already exist."""
+        try:
+            with self._driver.session(database="default") as session:
+                graph_list = session.run("CALL dbms.graph.listGraphs()").data()
+                exists = any(item["graph_name"] == graph_name for item in graph_list)
+        except Exception as e:
+            raise Exception(
+                f"Failed to check graph exist'{graph_name}': {str(e)}"
+            ) from e
+
+        return exists
+
     def delete_graph(self, graph_name: str) -> None:
         """Delete a graph in the database if it exists."""
         with self._driver.session(database="default") as session:

dbgpt/storage/graph_store/base.py

@@ -40,3 +40,7 @@ class GraphStoreBase(ABC):
     @abstractmethod
     def get_config(self) -> GraphStoreConfig:
         """Get the graph store config."""
+
+    def is_exist(self, name) -> bool:
+        """Check Graph Name is Exist."""
+        raise NotImplementedError

dbgpt/storage/graph_store/tugraph_store.py

@@ -102,6 +102,10 @@ class TuGraphStore(GraphStoreBase):
         """Get the TuGraph store config."""
         return self._config
 
+    def is_exist(self, name) -> bool:
+        """Check Graph Name is Exist."""
+        return self.conn.is_exist(name)
+
     def _add_vertex_index(self, field_name):
         """Add an index to the vertex table."""
         # TODO: Not used in the current implementation.

dbgpt/storage/knowledge_graph/community_summary.py

@@ -316,6 +316,8 @@ class CommunitySummaryKnowledgeGraph(BuiltinKnowledgeGraph):
 
     async def aload_document(self, chunks: List[Chunk]) -> List[str]:
         """Extract and persist graph from the document file."""
+        if not self.vector_name_exists():
+            self._graph_store_apdater.create_graph(self.get_config().name)
         await self._aload_document_graph(chunks)
         await self._aload_triplet_graph(chunks)
         await self._community_store.build_communities(

dbgpt/storage/knowledge_graph/knowledge_graph.py

@@ -172,6 +172,8 @@ class BuiltinKnowledgeGraph(KnowledgeGraphBase):
             return chunk.chunk_id
 
         # wait async tasks completed
+        if not self.vector_name_exists():
+            self._graph_store_apdater.create_graph(self.get_config().name)
         tasks = [process_chunk(chunk) for chunk in chunks]
         loop = asyncio.new_event_loop()
         asyncio.set_event_loop(loop)
@@ -187,6 +189,8 @@ class BuiltinKnowledgeGraph(KnowledgeGraphBase):
         Return:
             List[str]: chunk ids.
         """
+        if not self.vector_name_exists():
+            self._graph_store_apdater.create_graph(self.get_config().name)
         for chunk in chunks:
             triplets = await self._triplet_extractor.extract(chunk.content)
             for triplet in triplets:
@@ -279,3 +283,7 @@ class BuiltinKnowledgeGraph(KnowledgeGraphBase):
         """Delete by ids."""
         self._graph_store_apdater.delete_document(chunk_id=ids)
         return []
+
+    def vector_name_exists(self) -> bool:
+        """Whether name exists."""
+        return self._graph_store_apdater.graph_store.is_exist(self._config.name)