chore: add abstract decorator and output log when query embedding fails

api/core/embedding/cached_embedding.py

@@ -5,6 +5,7 @@
 import numpy as np
 from sqlalchemy.exc import IntegrityError
 
+from configs import dify_config
 from core.embedding.embedding_constant import EmbeddingInputType
 from core.model_manager import ModelInstance
 from core.model_runtime.entities.model_entities import ModelPropertyKey
@@ -110,6 +111,8 @@ def embed_query(self, text: str) -> list[float]:
             embedding_results = embedding_result.embeddings[0]
             embedding_results = (embedding_results / np.linalg.norm(embedding_results)).tolist()
         except Exception as ex:
+            if dify_config.DEBUG:
+                logging.exception(f"Failed to embed query text: {ex}")
             raise ex
 
         try:
@@ -122,6 +125,8 @@ def embed_query(self, text: str) -> list[float]:
             encoded_str = encoded_vector.decode("utf-8")
             redis_client.setex(embedding_cache_key, 600, encoded_str)
         except Exception as ex:
-            logging.exception("Failed to add embedding to redis %s", ex)
+            if dify_config.DEBUG:
+                logging.exception("Failed to add embedding to redis %s", ex)
+            raise ex
 
         return embedding_results

api/core/rag/datasource/keyword/keyword_base.py

@@ -27,9 +27,11 @@ def text_exists(self, id: str) -> bool:
     def delete_by_ids(self, ids: list[str]) -> None:
         raise NotImplementedError
 
+    @abstractmethod
     def delete(self) -> None:
         raise NotImplementedError
 
+    @abstractmethod
     def search(self, query: str, **kwargs: Any) -> list[Document]:
         raise NotImplementedError
 

api/core/rag/datasource/vdb/elasticsearch/elasticsearch_vector.py

@@ -77,7 +77,7 @@ def _check_version(self):
             raise ValueError("Elasticsearch vector database version must be greater than 8.0.0")
 
     def get_type(self) -> str:
-        return "elasticsearch"
+        return VectorType.ELASTICSEARCH
 
     def add_texts(self, documents: list[Document], embeddings: list[list[float]], **kwargs):
         uuids = self._get_uuids(documents)