fix: dataset query created_by empty UUID in iteration subgraph (#34004) (#34044)

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

api/core/app/apps/advanced_chat/app_runner.py

@@ -137,6 +137,7 @@ class AdvancedChatAppRunner(WorkflowBasedAppRunner):
                 workflow=self._workflow,
                 single_iteration_run=self.application_generate_entity.single_iteration_run,
                 single_loop_run=self.application_generate_entity.single_loop_run,
+                user_id=self.application_generate_entity.user_id,
             )
         else:
             inputs = self.application_generate_entity.inputs

api/core/app/apps/pipeline/pipeline_runner.py

@@ -106,6 +106,7 @@ class PipelineRunner(WorkflowBasedAppRunner):
                 workflow=workflow,
                 single_iteration_run=self.application_generate_entity.single_iteration_run,
                 single_loop_run=self.application_generate_entity.single_loop_run,
+                user_id=self.application_generate_entity.user_id,
             )
         else:
             inputs = self.application_generate_entity.inputs

api/core/app/apps/workflow/app_runner.py

@@ -92,6 +92,7 @@ class WorkflowAppRunner(WorkflowBasedAppRunner):
                 workflow=self._workflow,
                 single_iteration_run=self.application_generate_entity.single_iteration_run,
                 single_loop_run=self.application_generate_entity.single_loop_run,
+                user_id=self.application_generate_entity.user_id,
             )
         else:
             inputs = self.application_generate_entity.inputs

api/core/app/apps/workflow_app_runner.py

@@ -164,6 +164,8 @@ class WorkflowBasedAppRunner:
         workflow: Workflow,
         single_iteration_run: Any | None = None,
         single_loop_run: Any | None = None,
+        *,
+        user_id: str,
     ) -> tuple[Graph, VariablePool, GraphRuntimeState]:
         """
         Prepare graph, variable pool, and runtime state for single node execution
@@ -200,6 +202,7 @@ class WorkflowBasedAppRunner:
                 graph_runtime_state=graph_runtime_state,
                 node_type_filter_key="iteration_id",
                 node_type_label="iteration",
+                user_id=user_id,
             )
         elif single_loop_run:
             graph, variable_pool = self._get_graph_and_variable_pool_for_single_node_run(
@@ -209,6 +212,7 @@ class WorkflowBasedAppRunner:
                 graph_runtime_state=graph_runtime_state,
                 node_type_filter_key="loop_id",
                 node_type_label="loop",
+                user_id=user_id,
             )
         else:
             raise ValueError("Neither single_iteration_run nor single_loop_run is specified")
@@ -225,6 +229,8 @@ class WorkflowBasedAppRunner:
         graph_runtime_state: GraphRuntimeState,
         node_type_filter_key: str,  # 'iteration_id' or 'loop_id'
         node_type_label: str = "node",  # 'iteration' or 'loop' for error messages
+        *,
+        user_id: str = "",
     ) -> tuple[Graph, VariablePool]:
         """
         Get graph and variable pool for single node execution (iteration or loop).
@@ -290,7 +296,7 @@ class WorkflowBasedAppRunner:
             run_context=build_dify_run_context(
                 tenant_id=workflow.tenant_id,
                 app_id=self._app_id,
-                user_id="",
+                user_id=user_id,
                 user_from=UserFrom.ACCOUNT,
                 invoke_from=InvokeFrom.DEBUGGER,
             ),

api/core/rag/retrieval/dataset_retrieval.py

@@ -71,6 +71,7 @@ from graphon.model_runtime.entities.llm_entities import LLMMode, LLMResult, LLMU
 from graphon.model_runtime.entities.message_entities import PromptMessage, PromptMessageRole, PromptMessageTool
 from graphon.model_runtime.entities.model_entities import ModelFeature, ModelType
 from graphon.model_runtime.model_providers.__base.large_language_model import LargeLanguageModel
+from libs.helper import parse_uuid_str_or_none
 from libs.json_in_md_parser import parse_and_check_json_markdown
 from models import UploadFile
 from models.dataset import (
@@ -1024,8 +1025,13 @@ class DatasetRetrieval:
         """
         if not query and not attachment_ids:
             return
-        created_by_role = self._resolve_creator_user_role(user_from)
-        if created_by_role is None:
+        created_by = parse_uuid_str_or_none(user_id)
+        if created_by is None:
+            logger.debug(
+                "Skipping dataset query log: empty created_by user_id (user_from=%s, app_id=%s)",
+                user_from,
+                app_id,
+            )
             return
         dataset_queries = []
         for dataset_id in dataset_ids:
@@ -1041,8 +1047,8 @@ class DatasetRetrieval:
                     content=json.dumps(contents),
                     source=DatasetQuerySource.APP,
                     source_app_id=app_id,
-                    created_by_role=created_by_role,
-                    created_by=user_id,
+                    created_by_role=CreatorUserRole(user_from),
+                    created_by=created_by,
                 )
                 dataset_queries.append(dataset_query)
             if dataset_queries:

api/libs/helper.py

@@ -174,6 +174,18 @@ def normalize_uuid(value: str | UUID) -> str:
         raise ValueError("must be a valid UUID") from exc
 
 
+def parse_uuid_str_or_none(value: str | None) -> str | None:
+    """
+    Return None for missing/empty UUID-like values.
+
+    Keep non-empty values unchanged to avoid changing behavior in paths that
+    currently pass placeholder IDs in tests/mocks.
+    """
+    if value is None or not str(value).strip():
+        return None
+    return str(value)
+
+
 UUIDStrOrEmpty = Annotated[str, AfterValidator(normalize_uuid)]
 
 

api/tests/unit_tests/core/app/apps/test_workflow_app_runner_core.py

@@ -88,7 +88,7 @@ class TestWorkflowBasedAppRunner:
         workflow = SimpleNamespace(environment_variables=[], graph_dict={})
 
         with pytest.raises(ValueError, match="Neither single_iteration_run nor single_loop_run"):
-            runner._prepare_single_node_execution(workflow, None, None)
+            runner._prepare_single_node_execution(workflow, None, None, user_id="00000000-0000-0000-0000-000000000001")
 
     def test_get_graph_and_variable_pool_for_single_node_run(self, monkeypatch):
         runner = WorkflowBasedAppRunner(queue_manager=SimpleNamespace(), app_id="app")
@@ -136,6 +136,7 @@ class TestWorkflowBasedAppRunner:
             graph_runtime_state=graph_runtime_state,
             node_type_filter_key="iteration_id",
             node_type_label="iteration",
+            user_id="00000000-0000-0000-0000-000000000001",
         )
 
         assert graph is not None

api/tests/unit_tests/core/app/apps/test_workflow_app_runner_single_node.py

@@ -100,6 +100,7 @@ def test_run_uses_single_node_execution_branch(
         workflow=workflow,
         single_iteration_run=single_iteration_run,
         single_loop_run=single_loop_run,
+        user_id="user",
     )
     init_graph.assert_not_called()
 
@@ -158,6 +159,7 @@ def test_single_node_run_validates_target_node_config(monkeypatch) -> None:
             graph_runtime_state=graph_runtime_state,
             node_type_filter_key="loop_id",
             node_type_label="loop",
+            user_id="00000000-0000-0000-0000-000000000001",
         )
 
     assert seen_configs == [workflow.graph_dict["nodes"][0]]