chore: add warning for no new entities (#11)

nano_graphrag/_op.py

@@ -1,6 +1,7 @@
 import asyncio
 import json
 import re
+from typing import Union
 from collections import Counter, defaultdict
 
 from openai import AsyncOpenAI
@@ -223,7 +224,7 @@ async def extract_entities(
     knwoledge_graph_inst: BaseGraphStorage,
     entity_vdb: BaseVectorStorage,
     global_config: dict,
-) -> BaseGraphStorage:
+) -> Union[BaseGraphStorage, None]:
     use_llm_func: callable = global_config["best_model_func"]
     entity_extract_max_gleaning = global_config["entity_extract_max_gleaning"]
 
@@ -325,6 +326,9 @@ async def _process_single_content(chunk_key_dp: tuple[str, TextChunkSchema]):
             for k, v in maybe_edges.items()
         ]
     )
+    if not len(all_entities_data):
+        logger.warning("Didn't extract any entities, maybe your LLM is not working")
+        return None
     if entity_vdb is not None:
         data_for_vdb = {
             compute_mdhash_id(dp["entity_name"], prefix="ent-"): {

nano_graphrag/_storage.py

@@ -76,6 +76,9 @@ def __post_init__(self):
 
     async def upsert(self, data: dict[str, dict]):
         logger.info(f"Inserting {len(data)} vectors to {self.namespace}")
+        if not len(data):
+            logger.warning("You insert an empty data to vector DB")
+            return []
         list_data = [
             {
                 "__id__": k,

nano_graphrag/graphrag.py

@@ -228,13 +228,16 @@ async def ainsert(self, string_or_strings):
 
             # ---------- extract/summary entity and upsert to graph
             logger.info("[Entity Extraction]...")
-            self.chunk_entity_relation_graph = await extract_entities(
+            maybe_new_kg = await extract_entities(
                 inserting_chunks,
                 knwoledge_graph_inst=self.chunk_entity_relation_graph,
                 entity_vdb=self.entities_vdb,
                 global_config=asdict(self),
             )
-
+            if maybe_new_kg is None:
+                logger.warning("No new entities found")
+                return
+            self.chunk_entity_relation_graph = maybe_new_kg
             # ---------- update clusterings of graph
             logger.info("[Community Report]...")
             await self.chunk_entity_relation_graph.clustering(