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

api/core/ops/entities/config_entity.py

@@ -98,6 +98,7 @@ class WeaveConfig(BaseTracingConfig):
     entity: str | None = None
     project: str
     endpoint: str = "https://trace.wandb.ai"
+    host: str | None = None
 
     @field_validator("endpoint")
     @classmethod
@@ -109,6 +110,14 @@ class WeaveConfig(BaseTracingConfig):
 
         return v
 
+    @field_validator("host")
+    @classmethod
+    def validate_host(cls, v, info: ValidationInfo):
+        if v is not None and v != "":
+            if not v.startswith(("https://", "http://")):
+                raise ValueError("host must start with https:// or http://")
+        return v
+
 
 OPS_FILE_PATH = "ops_trace/"
 OPS_TRACE_FAILED_KEY = "FAILED_OPS_TRACE"

api/core/ops/ops_trace_manager.py

@@ -81,7 +81,7 @@ class OpsTraceProviderConfigMap(dict[str, dict[str, Any]]):
                 return {
                     "config_class": WeaveConfig,
                     "secret_keys": ["api_key"],
-                    "other_keys": ["project", "entity", "endpoint"],
+                    "other_keys": ["project", "entity", "endpoint", "host"],
                     "trace_instance": WeaveDataTrace,
                 }
 

api/core/ops/weave_trace/weave_trace.py

@@ -40,9 +40,14 @@ class WeaveDataTrace(BaseTraceInstance):
         self.weave_api_key = weave_config.api_key
         self.project_name = weave_config.project
         self.entity = weave_config.entity
+        self.host = weave_config.host
+
+        # Login with API key first, including host if provided
+        if self.host:
+            login_status = wandb.login(key=self.weave_api_key, verify=True, relogin=True, host=self.host)
+        else:
+            login_status = wandb.login(key=self.weave_api_key, verify=True, relogin=True)
 
-        # Login with API key first
-        login_status = wandb.login(key=self.weave_api_key, verify=True, relogin=True)
         if not login_status:
             logger.error("Failed to login to Weights & Biases with the provided API key")
             raise ValueError("Weave login failed")
@@ -386,7 +391,11 @@ class WeaveDataTrace(BaseTraceInstance):
 
     def api_check(self):
         try:
-            login_status = wandb.login(key=self.weave_api_key, verify=True, relogin=True)
+            if self.host:
+                login_status = wandb.login(key=self.weave_api_key, verify=True, relogin=True, host=self.host)
+            else:
+                login_status = wandb.login(key=self.weave_api_key, verify=True, relogin=True)
+
             if not login_status:
                 raise ValueError("Weave login failed")
             else:

api/core/rag/datasource/vdb/opensearch/opensearch_vector.py

@@ -184,7 +184,16 @@ class OpenSearchVector(BaseVector):
         }
         document_ids_filter = kwargs.get("document_ids_filter")
         if document_ids_filter:
-            query["query"] = {"terms": {"metadata.document_id": document_ids_filter}}
+            query["query"] = {
+                "script_score": {
+                    "query": {"bool": {"filter": [{"terms": {Field.DOCUMENT_ID.value: document_ids_filter}}]}},
+                    "script": {
+                        "source": "knn_score",
+                        "lang": "knn",
+                        "params": {"field": Field.VECTOR.value, "query_value": query_vector, "space_type": "l2"},
+                    },
+                }
+            }
 
         try:
             response = self._client.search(index=self._collection_name.lower(), body=query)

api/core/rag/datasource/vdb/oracle/oraclevector.py

@@ -303,7 +303,6 @@ class OracleVector(BaseVector):
             return docs
         else:
             return [Document(page_content="", metadata={})]
-        return []
 
     def delete(self) -> None:
         with self._get_connection() as conn:

api/core/tools/utils/dataset_retriever/dataset_multi_retriever_tool.py

@@ -153,8 +153,6 @@ class DatasetMultiRetrieverTool(DatasetRetrieverBaseTool):
             return str("\n".join(document_context_list))
         return ""
 
-        raise RuntimeError("not segments found")
-
     def _retriever(
         self,
         flask_app: Flask,

api/core/workflow/nodes/knowledge_retrieval/entities.py

@@ -128,3 +128,12 @@ class KnowledgeRetrievalNodeData(BaseNodeData):
     metadata_model_config: Optional[ModelConfig] = None
     metadata_filtering_conditions: Optional[MetadataFilteringCondition] = None
     vision: VisionConfig = Field(default_factory=VisionConfig)
+
+    @property
+    def structured_output_enabled(self) -> bool:
+        # NOTE(QuantumGhost): Temporary workaround for issue #20725
+        # (https://github.com/langgenius/dify/issues/20725).
+        #
+        # The proper fix would be to make `KnowledgeRetrievalNode` inherit
+        # from `BaseNode` instead of `LLMNode`.
+        return False

api/core/workflow/nodes/question_classifier/entities.py

@@ -19,3 +19,12 @@ class QuestionClassifierNodeData(BaseNodeData):
     instruction: Optional[str] = None
     memory: Optional[MemoryConfig] = None
     vision: VisionConfig = Field(default_factory=VisionConfig)
+
+    @property
+    def structured_output_enabled(self) -> bool:
+        # NOTE(QuantumGhost): Temporary workaround for issue #20725
+        # (https://github.com/langgenius/dify/issues/20725).
+        #
+        # The proper fix would be to make `QuestionClassifierNode` inherit
+        # from `BaseNode` instead of `LLMNode`.
+        return False