fix(workflow_service): assign UUID to workflow_node_execution id and update optional fields in WorkflowRun and WorkflowNodeExecution models (#12096)

Signed-off-by: -LAN- <[email protected]>

Author: laipz8200

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

@@ -101,10 +101,12 @@ def __init__(
         )
 
         if isinstance(user, EndUser):
-            self._user_id = user.session_id
+            self._user_id = user.id
+            user_session_id = user.session_id
             self._created_by_role = CreatedByRole.END_USER
         elif isinstance(user, Account):
             self._user_id = user.id
+            user_session_id = user.id
             self._created_by_role = CreatedByRole.ACCOUNT
         else:
             raise NotImplementedError(f"User type not supported: {type(user)}")
@@ -122,7 +124,7 @@ def __init__(
             SystemVariableKey.QUERY: message.query,
             SystemVariableKey.FILES: application_generate_entity.files,
             SystemVariableKey.CONVERSATION_ID: conversation.id,
-            SystemVariableKey.USER_ID: self._user_id,
+            SystemVariableKey.USER_ID: user_session_id,
             SystemVariableKey.DIALOGUE_COUNT: dialogue_count,
             SystemVariableKey.APP_ID: application_generate_entity.app_config.app_id,
             SystemVariableKey.WORKFLOW_ID: workflow.id,
@@ -134,6 +136,7 @@ def __init__(
 
         self._conversation_name_generate_thread = None
         self._recorded_files: list[Mapping[str, Any]] = []
+        self._workflow_run_id = ""
 
     def process(self):
         """
@@ -262,8 +265,7 @@ def _process_stream_response(
         :return:
         """
         # init fake graph runtime state
-        graph_runtime_state = None
-        workflow_run = None
+        graph_runtime_state: Optional[GraphRuntimeState] = None
 
         for queue_message in self._queue_manager.listen():
             event = queue_message.event
@@ -288,111 +290,163 @@ def _process_stream_response(
                         user_id=self._user_id,
                         created_by_role=self._created_by_role,
                     )
+                    self._workflow_run_id = workflow_run.id
                     message = self._get_message(session=session)
                     if not message:
                         raise ValueError(f"Message not found: {self._message_id}")
                     message.workflow_run_id = workflow_run.id
-                    session.commit()
-
                     workflow_start_resp = self._workflow_start_to_stream_response(
                         session=session, task_id=self._application_generate_entity.task_id, workflow_run=workflow_run
                     )
+                    session.commit()
+
                 yield workflow_start_resp
             elif isinstance(
                 event,
                 QueueNodeRetryEvent,
             ):
-                if not workflow_run:
+                if not self._workflow_run_id:
                     raise ValueError("workflow run not initialized.")
-                workflow_node_execution = self._handle_workflow_node_execution_retried(
-                    workflow_run=workflow_run, event=event
-                )
 
-                node_retry_resp = self._workflow_node_retry_to_stream_response(
-                    event=event,
-                    task_id=self._application_generate_entity.task_id,
-                    workflow_node_execution=workflow_node_execution,
-                )
+                with Session(db.engine) as session:
+                    workflow_run = self._get_workflow_run(session=session, workflow_run_id=self._workflow_run_id)
+                    workflow_node_execution = self._handle_workflow_node_execution_retried(
+                        session=session, workflow_run=workflow_run, event=event
+                    )
+                    node_retry_resp = self._workflow_node_retry_to_stream_response(
+                        session=session,
+                        event=event,
+                        task_id=self._application_generate_entity.task_id,
+                        workflow_node_execution=workflow_node_execution,
+                    )
+                    session.commit()
 
                 if node_retry_resp:
                     yield node_retry_resp
             elif isinstance(event, QueueNodeStartedEvent):
-                if not workflow_run:
+                if not self._workflow_run_id:
                     raise ValueError("workflow run not initialized.")
 
-                workflow_node_execution = self._handle_node_execution_start(workflow_run=workflow_run, event=event)
+                with Session(db.engine) as session:
+                    workflow_run = self._get_workflow_run(session=session, workflow_run_id=self._workflow_run_id)
+                    workflow_node_execution = self._handle_node_execution_start(
+                        session=session, workflow_run=workflow_run, event=event
+                    )
 
-                node_resp = self._workflow_node_start_to_stream_response(
-                    event=event,
-                    task_id=self._application_generate_entity.task_id,
-                    workflow_node_execution=workflow_node_execution,
-                )
+                    node_resp = self._workflow_node_start_to_stream_response(
+                        session=session,
+                        event=event,
+                        task_id=self._application_generate_entity.task_id,
+                        workflow_node_execution=workflow_node_execution,
+                    )
+                    session.commit()
 
                 if node_resp:
                     yield node_resp
             elif isinstance(event, QueueNodeSucceededEvent):
-                workflow_node_execution = self._handle_workflow_node_execution_success(event)
-
                 # Record files if it's an answer node or end node
                 if event.node_type in [NodeType.ANSWER, NodeType.END]:
                     self._recorded_files.extend(self._fetch_files_from_node_outputs(event.outputs or {}))
 
-                node_finish_resp = self._workflow_node_finish_to_stream_response(
-                    event=event,
-                    task_id=self._application_generate_entity.task_id,
-                    workflow_node_execution=workflow_node_execution,
-                )
+                with Session(db.engine) as session:
+                    workflow_node_execution = self._handle_workflow_node_execution_success(session=session, event=event)
+
+                    node_finish_resp = self._workflow_node_finish_to_stream_response(
+                        session=session,
+                        event=event,
+                        task_id=self._application_generate_entity.task_id,
+                        workflow_node_execution=workflow_node_execution,
+                    )
+                    session.commit()
 
                 if node_finish_resp:
                     yield node_finish_resp
             elif isinstance(event, QueueNodeFailedEvent | QueueNodeInIterationFailedEvent | QueueNodeExceptionEvent):
-                workflow_node_execution = self._handle_workflow_node_execution_failed(event)
+                with Session(db.engine) as session:
+                    workflow_node_execution = self._handle_workflow_node_execution_failed(session=session, event=event)
+
+                    node_finish_resp = self._workflow_node_finish_to_stream_response(
+                        session=session,
+                        event=event,
+                        task_id=self._application_generate_entity.task_id,
+                        workflow_node_execution=workflow_node_execution,
+                    )
+                    session.commit()
 
-                node_finish_resp = self._workflow_node_finish_to_stream_response(
-                    event=event,
-                    task_id=self._application_generate_entity.task_id,
-                    workflow_node_execution=workflow_node_execution,
-                )
                 if node_finish_resp:
                     yield node_finish_resp
-
             elif isinstance(event, QueueParallelBranchRunStartedEvent):
-                if not workflow_run:
+                if not self._workflow_run_id:
                     raise ValueError("workflow run not initialized.")
 
-                yield self._workflow_parallel_branch_start_to_stream_response(
-                    task_id=self._application_generate_entity.task_id, workflow_run=workflow_run, event=event
-                )
+                with Session(db.engine) as session:
+                    workflow_run = self._get_workflow_run(session=session, workflow_run_id=self._workflow_run_id)
+                    parallel_start_resp = self._workflow_parallel_branch_start_to_stream_response(
+                        session=session,
+                        task_id=self._application_generate_entity.task_id,
+                        workflow_run=workflow_run,
+                        event=event,
+                    )
+
+                yield parallel_start_resp
             elif isinstance(event, QueueParallelBranchRunSucceededEvent | QueueParallelBranchRunFailedEvent):
-                if not workflow_run:
+                if not self._workflow_run_id:
                     raise ValueError("workflow run not initialized.")
 
-                yield self._workflow_parallel_branch_finished_to_stream_response(
-                    task_id=self._application_generate_entity.task_id, workflow_run=workflow_run, event=event
-                )
+                with Session(db.engine) as session:
+                    workflow_run = self._get_workflow_run(session=session, workflow_run_id=self._workflow_run_id)
+                    parallel_finish_resp = self._workflow_parallel_branch_finished_to_stream_response(
+                        session=session,
+                        task_id=self._application_generate_entity.task_id,
+                        workflow_run=workflow_run,
+                        event=event,
+                    )
+
+                yield parallel_finish_resp
             elif isinstance(event, QueueIterationStartEvent):
-                if not workflow_run:
+                if not self._workflow_run_id:
                     raise ValueError("workflow run not initialized.")
 
-                yield self._workflow_iteration_start_to_stream_response(
-                    task_id=self._application_generate_entity.task_id, workflow_run=workflow_run, event=event
-                )
+                with Session(db.engine) as session:
+                    workflow_run = self._get_workflow_run(session=session, workflow_run_id=self._workflow_run_id)
+                    iter_start_resp = self._workflow_iteration_start_to_stream_response(
+                        session=session,
+                        task_id=self._application_generate_entity.task_id,
+                        workflow_run=workflow_run,
+                        event=event,
+                    )
+
+                yield iter_start_resp
             elif isinstance(event, QueueIterationNextEvent):
-                if not workflow_run:
+                if not self._workflow_run_id:
                     raise ValueError("workflow run not initialized.")
 
-                yield self._workflow_iteration_next_to_stream_response(
-                    task_id=self._application_generate_entity.task_id, workflow_run=workflow_run, event=event
-                )
+                with Session(db.engine) as session:
+                    workflow_run = self._get_workflow_run(session=session, workflow_run_id=self._workflow_run_id)
+                    iter_next_resp = self._workflow_iteration_next_to_stream_response(
+                        session=session,
+                        task_id=self._application_generate_entity.task_id,
+                        workflow_run=workflow_run,
+                        event=event,
+                    )
+
+                yield iter_next_resp
             elif isinstance(event, QueueIterationCompletedEvent):
-                if not workflow_run:
+                if not self._workflow_run_id:
                     raise ValueError("workflow run not initialized.")
 
-                yield self._workflow_iteration_completed_to_stream_response(
-                    task_id=self._application_generate_entity.task_id, workflow_run=workflow_run, event=event
-                )
+                with Session(db.engine) as session:
+                    workflow_run = self._get_workflow_run(session=session, workflow_run_id=self._workflow_run_id)
+                    iter_finish_resp = self._workflow_iteration_completed_to_stream_response(
+                        session=session,
+                        task_id=self._application_generate_entity.task_id,
+                        workflow_run=workflow_run,
+                        event=event,
+                    )
+
+                yield iter_finish_resp
             elif isinstance(event, QueueWorkflowSucceededEvent):
-                if not workflow_run:
+                if not self._workflow_run_id:
                     raise ValueError("workflow run not initialized.")
 
                 if not graph_runtime_state:
@@ -401,7 +455,7 @@ def _process_stream_response(
                 with Session(db.engine) as session:
                     workflow_run = self._handle_workflow_run_success(
                         session=session,
-                        workflow_run=workflow_run,
+                        workflow_run_id=self._workflow_run_id,
                         start_at=graph_runtime_state.start_at,
                         total_tokens=graph_runtime_state.total_tokens,
                         total_steps=graph_runtime_state.node_run_steps,
@@ -418,16 +472,15 @@ def _process_stream_response(
                 yield workflow_finish_resp
                 self._queue_manager.publish(QueueAdvancedChatMessageEndEvent(), PublishFrom.TASK_PIPELINE)
             elif isinstance(event, QueueWorkflowPartialSuccessEvent):
-                if not workflow_run:
+                if not self._workflow_run_id:
                     raise ValueError("workflow run not initialized.")
-
                 if not graph_runtime_state:
                     raise ValueError("graph runtime state not initialized.")
 
                 with Session(db.engine) as session:
                     workflow_run = self._handle_workflow_run_partial_success(
                         session=session,
-                        workflow_run=workflow_run,
+                        workflow_run_id=self._workflow_run_id,
                         start_at=graph_runtime_state.start_at,
                         total_tokens=graph_runtime_state.total_tokens,
                         total_steps=graph_runtime_state.node_run_steps,
@@ -436,7 +489,6 @@ def _process_stream_response(
                         conversation_id=None,
                         trace_manager=trace_manager,
                     )
-
                     workflow_finish_resp = self._workflow_finish_to_stream_response(
                         session=session, task_id=self._application_generate_entity.task_id, workflow_run=workflow_run
                     )
@@ -445,16 +497,15 @@ def _process_stream_response(
                 yield workflow_finish_resp
                 self._queue_manager.publish(QueueAdvancedChatMessageEndEvent(), PublishFrom.TASK_PIPELINE)
             elif isinstance(event, QueueWorkflowFailedEvent):
-                if not workflow_run:
+                if not self._workflow_run_id:
                     raise ValueError("workflow run not initialized.")
-
                 if not graph_runtime_state:
                     raise ValueError("graph runtime state not initialized.")
 
                 with Session(db.engine) as session:
                     workflow_run = self._handle_workflow_run_failed(
                         session=session,
-                        workflow_run=workflow_run,
+                        workflow_run_id=self._workflow_run_id,
                         start_at=graph_runtime_state.start_at,
                         total_tokens=graph_runtime_state.total_tokens,
                         total_steps=graph_runtime_state.node_run_steps,
@@ -470,15 +521,16 @@ def _process_stream_response(
                     err_event = QueueErrorEvent(error=ValueError(f"Run failed: {workflow_run.error}"))
                     err = self._handle_error(event=err_event, session=session, message_id=self._message_id)
                     session.commit()
+
                 yield workflow_finish_resp
                 yield self._error_to_stream_response(err)
                 break
             elif isinstance(event, QueueStopEvent):
-                if workflow_run and graph_runtime_state:
+                if self._workflow_run_id and graph_runtime_state:
                     with Session(db.engine) as session:
                         workflow_run = self._handle_workflow_run_failed(
                             session=session,
-                            workflow_run=workflow_run,
+                            workflow_run_id=self._workflow_run_id,
                             start_at=graph_runtime_state.start_at,
                             total_tokens=graph_runtime_state.total_tokens,
                             total_steps=graph_runtime_state.node_run_steps,
@@ -487,7 +539,6 @@ def _process_stream_response(
                             conversation_id=self._conversation_id,
                             trace_manager=trace_manager,
                         )
-
                         workflow_finish_resp = self._workflow_finish_to_stream_response(
                             session=session,
                             task_id=self._application_generate_entity.task_id,
@@ -496,6 +547,7 @@ def _process_stream_response(
                         # Save message
                         self._save_message(session=session, graph_runtime_state=graph_runtime_state)
                         session.commit()
+
                     yield workflow_finish_resp
 
                 yield self._message_end_to_stream_response()