.

scripts/rag_build_query.py

@@ -19,7 +19,7 @@ class BGEEmbedding(HuggingFaceEmbedding):
         prefix = "Represent this sentence for searching relevant passages: "
         return super()._get_query_embedding(prefix + query)
 
-    def _get_query_embeddings(self, queries: List[str]) -> List[List[float]]:
+    def _get_query_embeddings(self, queries: List[str]) -> List[List[float]]]:
         # 批量生成嵌入向量
         prefix = "Represent this sentence for searching relevant passages: "
         return super()._get_query_embeddings([prefix + q for q in queries])
@@ -89,13 +89,13 @@ def build_user_index(user_id: str):
     )
 
     try:
-        # 构建索引，并使用 `save()` 方法保存索引
+        # 构建索引，并使用 `storage_context.persist()` 方法保存索引
         index = VectorStoreIndex.from_documents(
             documents,
             service_context=service_context,
             storage_context=storage_context
         )
-        index.save(persist_dir=persist_dir)  # 使用 `save()` 方法代替 `persist()`
+        storage_context.persist(persist_dir=persist_dir)  # 使用 `storage_context.persist()` 保存索引
         logger.info(f"索引已保存到 {persist_dir}")
     except Exception as e:
         logger.error(f"索引构建失败: {e}")