feat: add GraphRAG framework and integrate TuGraph (#1506)

Co-authored-by: KingSkyLi <15566300566@163.com>
Co-authored-by: aries_ckt <916701291@qq.com>
Co-authored-by: Fangyin Cheng <staneyffer@gmail.com>

dbgpt/rag/assembler/base.py

@@ -1,4 +1,5 @@
 """Base Assembler."""
+
 from abc import ABC, abstractmethod
 from typing import Any, List, Optional
 
@@ -37,13 +38,15 @@ class BaseAssembler(ABC):
         )
         self._chunks: List[Chunk] = []
         metadata = {
-            "knowledge_cls": self._knowledge.__class__.__name__
-            if self._knowledge
-            else None,
+            "knowledge_cls": (
+                self._knowledge.__class__.__name__ if self._knowledge else None
+            ),
             "knowledge_type": self._knowledge.type().value if self._knowledge else None,
-            "path": self._knowledge._path
-            if self._knowledge and hasattr(self._knowledge, "_path")
-            else None,
+            "path": (
+                self._knowledge._path
+                if self._knowledge and hasattr(self._knowledge, "_path")
+                else None
+            ),
             "chunk_parameters": self._chunk_parameters.dict(),
         }
         with root_tracer.start_span("BaseAssembler.load_knowledge", metadata=metadata):
@@ -70,6 +73,14 @@ class BaseAssembler(ABC):
             List[str]: List of persisted chunk ids.
         """
 
+    async def apersist(self) -> List[str]:
+        """Persist chunks.
+
+        Returns:
+            List[str]: List of persisted chunk ids.
+        """
+        raise NotImplementedError
+
     def get_chunks(self) -> List[Chunk]:
         """Return chunks."""
         return self._chunks

dbgpt/rag/assembler/embedding.py

@@ -106,6 +106,14 @@ class EmbeddingAssembler(BaseAssembler):
         """
         return self._vector_store_connector.load_document(self._chunks)
 
+    async def apersist(self) -> List[str]:
+        """Persist chunks into store.
+
+        Returns:
+            List[str]: List of chunk ids.
+        """
+        return await self._vector_store_connector.aload_document(self._chunks)
+
     def _extract_info(self, chunks) -> List[Chunk]:
         """Extract info from chunks."""
         return []