Merge branch 'feat/rag-pipeline' into deploy/rag-dev

api/core/rag/embedding/cached_embedding.py

@@ -139,4 +139,4 @@ class CacheEmbedding(Embeddings):
                 logging.exception(f"Failed to add embedding to redis for the text '{text[:10]}...({len(text)} chars)'")
             raise ex
 
-        return embedding_results
+        return embedding_results  # type: ignore

api/core/rag/index_processor/processor/qa_index_processor.py

@@ -105,7 +105,7 @@ class QAIndexProcessor(BaseIndexProcessor):
 
     def format_by_template(self, file: FileStorage, **kwargs) -> list[Document]:
         # check file type
-        if not file.filename.endswith(".csv"):
+        if not file.filename or not file.filename.endswith(".csv"):
             raise ValueError("Invalid file type. Only CSV files are allowed")
 
         try:

api/core/rag/retrieval/router/multi_dataset_react_route.py

@@ -9,7 +9,7 @@ from core.prompt.advanced_prompt_transform import AdvancedPromptTransform
 from core.prompt.entities.advanced_prompt_entities import ChatModelMessage, CompletionModelPromptTemplate
 from core.rag.retrieval.output_parser.react_output import ReactAction
 from core.rag.retrieval.output_parser.structured_chat import StructuredChatOutputParser
-from core.workflow.nodes.llm import LLMNode
+from core.workflow.nodes.llm import llm_utils
 
 PREFIX = """Respond to the human as helpfully and accurately as possible. You have access to the following tools:"""
 
@@ -165,7 +165,7 @@ class ReactMultiDatasetRouter:
         text, usage = self._handle_invoke_result(invoke_result=invoke_result)
 
         # deduct quota
-        LLMNode.deduct_llm_quota(tenant_id=tenant_id, model_instance=model_instance, usage=usage)
+        llm_utils.deduct_llm_quota(tenant_id=tenant_id, model_instance=model_instance, usage=usage)
 
         return text, usage