chore(api): bump minimum supported Python to 3.13 (#43022)

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: Asuka Minato <i@asukaminato.eu.org>
Co-authored-by: Stephen Zhou <38493346+hyoban@users.noreply.github.com>

.devcontainer/Dockerfile

@@ -1,4 +1,4 @@
-FROM mcr.microsoft.com/devcontainers/python:3.12-bookworm
+FROM mcr.microsoft.com/devcontainers/python:3.13-bookworm
 
 RUN apt-get update && export DEBIAN_FRONTEND=noninteractive \
      && apt-get -y install libgmp-dev libmpfr-dev libmpc-dev

.devcontainer/devcontainer.json

@@ -1,7 +1,7 @@
 // For format details, see https://aka.ms/devcontainer.json. For config options, see the
 // README at: https://github.com/devcontainers/templates/tree/main/src/anaconda
 {
-	"name": "Python 3.12",
+	"name": "Python 3.13",
 	"build": {
 		"context": "..",
 		"dockerfile": "Dockerfile"

.github/workflows/accessibility-e2e.yml

@@ -57,7 +57,7 @@ jobs:
         uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
         with:
           enable-cache: true
-          python-version: '3.12'
+          python-version: '3.13'
           cache-dependency-glob: |
             api/uv.lock
 

.github/workflows/api-tests.yml

@@ -25,7 +25,7 @@ jobs:
       - name: Setup UV and Python
         uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
         with:
-          python-version: '3.12'
+          python-version: '3.13'
           enable-cache: false
       - name: Download shared timing baseline
         # Public statistics are independent of PR refs and require no credentials.
@@ -72,7 +72,7 @@ jobs:
       fail-fast: false
       matrix:
         python-version:
-          - '3.12'
+          - '3.13'
         shardIndex:
           - 1
           - 2
@@ -152,11 +152,11 @@ jobs:
           retention-days: 1
 
   api-controllers:
-    name: API Controller Tests (3.12)
+    name: API Controller Tests (3.13)
     runs-on: depot-ubuntu-24.04
     env:
       PYTHONUNBUFFERED: '1'
-      COVERAGE_FILE: coverage-controllers-3.12
+      COVERAGE_FILE: coverage-controllers-3.13
     steps:
       - name: Checkout code
         uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
@@ -167,7 +167,7 @@ jobs:
         uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
         with:
           enable-cache: true
-          python-version: '3.12'
+          python-version: '3.13'
           cache-dependency-glob: api/uv.lock
 
       - name: Check UV lockfile
@@ -192,7 +192,7 @@ jobs:
         if: ${{ !cancelled() }}
         uses: actions/upload-artifact@cf430e030ddbb5b0abf93d22962f4752f3646cd9 # v7.0.2
         with:
-          name: api-test-results-controllers-3.12
+          name: api-test-results-controllers-3.13
           path: test-results/*.xml
           if-no-files-found: warn
           retention-days: 7
@@ -200,7 +200,7 @@ jobs:
       - name: Upload controller coverage data
         uses: actions/upload-artifact@cf430e030ddbb5b0abf93d22962f4752f3646cd9 # v7.0.2
         with:
-          name: api-coverage-controllers-3.12
+          name: api-coverage-controllers-3.13
           path: ${{ env.COVERAGE_FILE }}
           if-no-files-found: error
           retention-days: 1
@@ -220,7 +220,7 @@ jobs:
       fail-fast: false
       matrix:
         python-version:
-          - '3.12'
+          - '3.13'
 
     steps:
       - name: Checkout code
@@ -286,7 +286,7 @@ jobs:
       - name: Setup UV and Python
         uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
         with:
-          python-version: '3.12'
+          python-version: '3.13'
 
       - name: Download coverage data
         uses: actions/download-artifact@9000827ccba6bdab643e8b6fd33ac0654aef8333 # v8.0.2

.github/workflows/autofix.yml

@@ -61,7 +61,7 @@ jobs:
       - if: github.event_name != 'merge_group' && (steps.docker-compose-changes.outputs.any_changed == 'true' || steps.api-changes.outputs.any_changed == 'true' || steps.frontend-contract-changes.outputs.any_changed == 'true' || steps.dify-agent-changes.outputs.any_changed == 'true')
         uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
         with:
-          python-version: '3.11'
+          python-version: '3.13'
 
       - if: github.event_name != 'merge_group' && (steps.api-changes.outputs.any_changed == 'true' || steps.frontend-contract-changes.outputs.any_changed == 'true' || steps.dify-agent-changes.outputs.any_changed == 'true')
         uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0

.github/workflows/db-migration-test.yml

@@ -22,7 +22,7 @@ jobs:
         uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
         with:
           enable-cache: true
-          python-version: '3.12'
+          python-version: '3.13'
           cache-dependency-glob: api/uv.lock
 
       - name: Install dependencies
@@ -72,7 +72,7 @@ jobs:
         uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
         with:
           enable-cache: true
-          python-version: '3.12'
+          python-version: '3.13'
           cache-dependency-glob: api/uv.lock
 
       - name: Install dependencies

.github/workflows/dify-agent-tests.yml

@@ -21,7 +21,7 @@ jobs:
     strategy:
       matrix:
         python-version:
-          - '3.12'
+          - '3.13'
 
     steps:
       - name: Checkout code

.github/workflows/style.yml

@@ -64,7 +64,7 @@ jobs:
         uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
         with:
           enable-cache: false
-          python-version: '3.12'
+          python-version: '3.13'
           cache-dependency-glob: |
             api/uv.lock
             dify-agent/uv.lock

.github/workflows/vdb-tests-full.yml

@@ -20,7 +20,7 @@ jobs:
     strategy:
       matrix:
         python-version:
-          - '3.12'
+          - '3.13'
 
     steps:
       - name: Checkout code

.github/workflows/vdb-tests.yml

@@ -17,7 +17,7 @@ jobs:
     strategy:
       matrix:
         python-version:
-          - '3.12'
+          - '3.13'
 
     steps:
       - name: Checkout code

.github/workflows/web-e2e.yml

@@ -118,7 +118,7 @@ jobs:
         uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
         with:
           enable-cache: true
-          python-version: '3.12'
+          python-version: '3.13'
           cache-dependency-glob: api/uv.lock
 
       # API and web installs are independent, so overlap them; the browser
@@ -196,7 +196,7 @@ jobs:
         uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
         with:
           enable-cache: true
-          python-version: '3.12'
+          python-version: '3.13'
           cache-dependency-glob: api/uv.lock
 
       # API and web installs are independent, so overlap them; the browser
@@ -268,7 +268,7 @@ jobs:
         uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
         with:
           enable-cache: true
-          python-version: '3.12'
+          python-version: '3.13'
           cache-dependency-glob: |
             api/uv.lock
             dify-agent/uv.lock

api/Dockerfile

@@ -1,5 +1,5 @@
 # base image
-FROM python:3.12-slim-bookworm AS base
+FROM python:3.13-slim-bookworm AS base
 
 WORKDIR /app/api
 

api/context/execution_context.py

@@ -94,7 +94,7 @@ class ExecutionContext:
         return self._user
 
     @contextmanager
-    def enter(self) -> Generator[None, None, None]:
+    def enter(self) -> Generator[None]:
         """Enter this execution context."""
         if self._context_vars:
             for var, val in self._context_vars.items():
@@ -145,7 +145,7 @@ class NullAppContext(AppContext):
 
     @contextmanager
     @override
-    def enter(self) -> Generator[None, None, None]:
+    def enter(self) -> Generator[None]:
         """Enter null context (no-op)."""
         yield
 

api/context/flask_app_context.py

@@ -42,7 +42,7 @@ class FlaskAppContext(AppContext):
 
     @contextmanager
     @override
-    def enter(self) -> Generator[None, None, None]:
+    def enter(self) -> Generator[None]:
         """Enter Flask app context."""
         with self._flask_app.app_context():
             yield
@@ -159,7 +159,7 @@ class FlaskExecutionContext:
             cm.__exit__(*args)
 
     @contextmanager
-    def enter(self) -> Generator[None, None, None]:
+    def enter(self) -> Generator[None]:
         """Enter Flask execution context as context manager."""
         # Restore non-Flask context variables to avoid leaking Flask tokens across threads
         for var, val in self._context_vars.items():

api/controllers/common/resource_access_token_errors.py

@@ -14,7 +14,7 @@ from services.auth.resource_access_token_contracts import (
 
 
 @contextmanager
-def resource_access_token_errors() -> Generator[None, None, None]:
+def resource_access_token_errors() -> Generator[None]:
     try:
         yield
     except ResourceAccessTokenInputError as error:

api/controllers/console/app/completion.py

@@ -473,7 +473,7 @@ def _create_build_chat_finalization_message(
     return {"result": "success"}, 200
 
 
-def _drain_streaming_generate_response(response: RateLimitGenerator | Generator[str, None, None]) -> None:
+def _drain_streaming_generate_response(response: RateLimitGenerator | Generator[str]) -> None:
     """Consume a streamed app-generate response until a terminal message event arrives.
 
     Finalize keeps the normal Agent App streaming path so the existing queue,
@@ -631,7 +631,7 @@ def _raise_agent_stream_error_before_response(response):
         return _prepend_stream_chunks(buffered, chunk, iterator)
 
 
-def _prepend_stream_chunks(buffered: list[Any], first: Any, iterator: Iterator[Any]) -> Generator[Any, None, None]:
+def _prepend_stream_chunks(buffered: list[Any], first: Any, iterator: Iterator[Any]) -> Generator[Any]:
     yield from buffered
     yield first
     yield from iterator

api/controllers/console/app/generator.py

@@ -602,7 +602,7 @@ class WorkflowGenerateStreamApi(Resource):
         if guard is not None:
             return guard
 
-        def generate() -> Generator[str, None, None]:
+        def generate() -> Generator[str]:
             try:
                 for event_name, payload in WorkflowGeneratorService.generate_workflow_graph_stream(
                     tenant_id=current_tenant_id,

api/controllers/console/files.py

@@ -85,7 +85,7 @@ def _parse_file_upload() -> tuple[FileStorage, str, Literal["datasets"] | None]:
 
 
 @contextmanager
-def _file_upload_errors() -> Generator[None, None, None]:
+def _file_upload_errors() -> Generator[None]:
     try:
         yield
     except services.errors.file.FileTooLargeError as file_too_large_error:

api/controllers/console/human_input_form.py

@@ -198,7 +198,7 @@ class ConsoleWorkflowEventsApi(Resource):
             payload = response.model_dump(mode="json")
             payload["event"] = response.event.value
 
-            def _generate_finished_events() -> Generator[str, None, None]:
+            def _generate_finished_events() -> Generator[str]:
                 yield f"data: {json.dumps(payload)}\n\n"
 
             event_generator = _generate_finished_events

api/controllers/inner_api/agent/llm.py

@@ -73,7 +73,7 @@ class AgentLLMInvokeApi(Resource):
                 status_code=400,
             ) from exc
 
-        def generate() -> Generator[str, None, None]:
+        def generate() -> Generator[str]:
             try:
                 for chunk in service.invoke(prepared):
                     envelope = {"code": 0, "message": "", "data": chunk.model_dump(mode="json")}

api/controllers/openapi/_hints.py

@@ -62,7 +62,7 @@ def _wanted_event(chunk: str, event: str) -> dict[str, Any] | None:
     return parsed
 
 
-def attach_stream_hints(events: Iterable[str], *, event: str, build: HintBuilder) -> Generator[str, None, None]:
+def attach_stream_hints(events: Iterable[str], *, event: str, build: HintBuilder) -> Generator[str]:
     """Yield the source SSE chunks, adding a top-level `hints` list to every `event` that `build` hints.
 
     `event: ping` chunks, other event kinds, events `build` returns nothing for, and events

api/controllers/openapi/app_run.py

@@ -91,7 +91,7 @@ _INVALID_RUN_INPUT: Final = "invalid run input"
 
 
 @contextmanager
-def _translate_service_errors() -> Generator[None, None, None]:
+def _translate_service_errors() -> Generator[None]:
     try:
         yield
     except WorkflowNotFoundError as ex:
@@ -191,7 +191,7 @@ class _ChatMessageEnd(MessageEndStreamResponse):
     conversation_id: str
 
 
-def with_reply_hints(events: Iterable[str], *, op: str, app_id: str) -> Generator[str, None, None]:
+def with_reply_hints(events: Iterable[str], *, op: str, app_id: str) -> Generator[str]:
     def build(event: Mapping[str, Any]) -> list[Hint]:
         end = _ChatMessageEnd.model_validate(event)
         return [
@@ -205,14 +205,14 @@ def with_reply_hints(events: Iterable[str], *, op: str, app_id: str) -> Generato
     return attach_stream_hints(events, event=StreamEvent.MESSAGE_END.value, build=build)
 
 
-HintLayer = Callable[[Iterable[str], str, str], Generator[str, None, None]]
+HintLayer = Callable[[Iterable[str], str, str], Generator[str]]
 
 
-def _reply_layer(events: Iterable[str], op: str, app_id: str) -> Generator[str, None, None]:
+def _reply_layer(events: Iterable[str], op: str, app_id: str) -> Generator[str]:
     return with_reply_hints(events, op=op, app_id=app_id)
 
 
-def _form_layer(events: Iterable[str], op: str, app_id: str) -> Generator[str, None, None]:
+def _form_layer(events: Iterable[str], op: str, app_id: str) -> Generator[str]:
     return with_form_hints(events, app_id=app_id)
 
 

api/controllers/openapi/human_input_form.py

@@ -219,7 +219,7 @@ def form_hints(*, op: str, app_id: str, response: HumanInputRequiredResponse) ->
     ]
 
 
-def with_form_hints(events: Iterable[str], *, app_id: str) -> Generator[str, None, None]:
+def with_form_hints(events: Iterable[str], *, app_id: str) -> Generator[str]:
     """A run stream whose `human_input_required` events carry hints that target the submit route above."""
 
     op = op_of(OpenApiWorkflowHumanInputFormSubmitApi.post)

api/controllers/openapi/workflow_events.py

@@ -118,7 +118,7 @@ class OpenApiWorkflowEventsApi(Resource):
             payload = response.model_dump(mode="json")
             payload["event"] = response.event.value
 
-            def _generate_finished_events() -> Generator[str, None, None]:
+            def _generate_finished_events() -> Generator[str]:
                 yield f"data: {json.dumps(payload)}\n\n"
 
             event_generator = _generate_finished_events

api/controllers/service_api/app/workflow_events.py

@@ -128,7 +128,7 @@ class WorkflowEventsApi(Resource):
             payload = response.model_dump(mode="json")
             payload["event"] = response.event.value
 
-            def _generate_finished_events() -> Generator[str, None, None]:
+            def _generate_finished_events() -> Generator[str]:
                 yield f"data: {json.dumps(payload)}\n\n"
 
             event_generator = _generate_finished_events

api/controllers/service_api/dataset/rag_pipeline/rag_pipeline_workflow.py

@@ -315,7 +315,7 @@ class PipelineRunApi(DatasetApiResource):
         rag_pipeline_service = RagPipelineService(session)
         pipeline = rag_pipeline_service.get_pipeline(tenant_id=tenant_id, dataset_id=dataset_id_str)
         try:
-            response: dict[Any, Any] | Generator[str, Any, None] = PipelineGenerateService.generate(
+            response: dict[Any, Any] | Generator[str, Any] = PipelineGenerateService.generate(
                 generator=application_services().knowledge.pipeline_generator,
                 session=session,
                 pipeline=pipeline,

api/controllers/web/workflow_events.py

@@ -69,7 +69,7 @@ class WorkflowEventsApi(WebApiResource):
             payload = response.model_dump(mode="json")
             payload["event"] = response.event.value
 
-            def _generate_finished_events() -> Generator[str, None, None]:
+            def _generate_finished_events() -> Generator[str]:
                 yield f"data: {json.dumps(payload)}\n\n"
 
             event_generator = _generate_finished_events

api/core/agent/fc_agent_runner.py

@@ -100,9 +100,7 @@ class FunctionCallAgentRunner(BaseAgentRunner):
 
         return prompt_message_contents
 
-    def run(
-        self, session: Session, message: Message, query: str, **kwargs: Any
-    ) -> Generator[LLMResultChunk, None, None]:
+    def run(self, session: Session, message: Message, query: str, **kwargs: Any) -> Generator[LLMResultChunk]:
         """
         Run FunctionCall agent application
         """
@@ -175,7 +173,7 @@ class FunctionCallAgentRunner(BaseAgentRunner):
                 "created_by": CreditUsageCreatedBy.APP,
             }
 
-            chunks: Union[Generator[LLMResultChunk, None, None], LLMResult] = model_instance.invoke_llm(
+            chunks: Union[Generator[LLMResultChunk], LLMResult] = model_instance.invoke_llm(
                 prompt_messages=prompt_messages,
                 model_parameters=app_generate_entity.model_conf.parameters,
                 tools=prompt_messages_tools,

api/core/agent/output_parser/cot_output_parser.py

@@ -10,8 +10,8 @@ from graphon.model_runtime.entities.llm_entities import LLMResultChunk
 class CotAgentOutputParser:
     @classmethod
     def handle_react_stream_output(
-        cls, llm_response: Generator[LLMResultChunk, None, None], usage_dict: dict[str, Any]
-    ) -> Generator[Union[str, AgentScratchpadUnit.Action], None, None]:
+        cls, llm_response: Generator[LLMResultChunk], usage_dict: dict[str, Any]
+    ) -> Generator[Union[str, AgentScratchpadUnit.Action]]:
         def parse_action(action) -> Union[str, AgentScratchpadUnit.Action]:
             action_name = None
             action_input = None

api/core/agent/strategy/base.py

@@ -20,7 +20,7 @@ class BaseAgentStrategy(ABC):
         app_id: str | None = None,
         message_id: str | None = None,
         credentials: InvokeCredentials | None = None,
-    ) -> Generator[AgentInvokeMessage, None, None]:
+    ) -> Generator[AgentInvokeMessage]:
         """
         Invoke the agent strategy.
         """
@@ -41,5 +41,5 @@ class BaseAgentStrategy(ABC):
         app_id: str | None = None,
         message_id: str | None = None,
         credentials: InvokeCredentials | None = None,
-    ) -> Generator[AgentInvokeMessage, None, None]:
+    ) -> Generator[AgentInvokeMessage]:
         pass

api/core/agent/strategy/plugin.py

@@ -44,7 +44,7 @@ class PluginAgentStrategy(BaseAgentStrategy):
         app_id: str | None = None,
         message_id: str | None = None,
         credentials: InvokeCredentials | None = None,
-    ) -> Generator[AgentInvokeMessage, None, None]:
+    ) -> Generator[AgentInvokeMessage]:
         """
         Invoke the agent strategy.
         """

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

@@ -103,7 +103,7 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
         pause_state_config: PauseStateLayerConfig | None = None,
         *,
         session: Session,
-    ) -> Generator[Mapping | str, None, None]: ...
+    ) -> Generator[Mapping | str]: ...
 
     @overload
     def generate(
@@ -118,7 +118,7 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
         pause_state_config: PauseStateLayerConfig | None = None,
         *,
         session: Session,
-    ) -> Mapping[str, Any] | Generator[str | Mapping, None, None]: ...
+    ) -> Mapping[str, Any] | Generator[str | Mapping]: ...
 
     def generate(
         self,
@@ -132,7 +132,7 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
         pause_state_config: PauseStateLayerConfig | None = None,
         *,
         session: Session,
-    ) -> Mapping[str, Any] | Generator[str | Mapping, None, None]:
+    ) -> Mapping[str, Any] | Generator[str | Mapping]:
         """
         Generate App response.
 
@@ -327,7 +327,7 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
         streaming: bool = True,
         *,
         session: Session,
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], None, None]:
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any]]:
         """
         Generate App response.
 
@@ -422,7 +422,7 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
         streaming: bool = True,
         *,
         session: Session,
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], None, None]:
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any]]:
         """
         Generate App response.
 
@@ -523,7 +523,7 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
         graph_runtime_state: GraphRuntimeState | None = None,
         graph_engine_layers: Sequence[GraphEngineLayer] = (),
         response_stream_filter: ResponseStreamFilter | None = None,
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], None, None]:
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any]]:
         """
         Generate App response.
 
@@ -760,11 +760,7 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
         user: Account | EndUser,
         draft_var_saver_factory: DraftVariableSaverFactory,
         stream: bool = False,
-    ) -> (
-        ChatbotAppBlockingResponse
-        | AdvancedChatPausedBlockingResponse
-        | Generator[ChatbotAppStreamResponse, None, None]
-    ):
+    ) -> ChatbotAppBlockingResponse | AdvancedChatPausedBlockingResponse | Generator[ChatbotAppStreamResponse]:
         """
         Handle response.
         :param application_generate_entity: application generate entity

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

@@ -80,8 +80,8 @@ class AdvancedChatAppGenerateResponseConverter(
     @classmethod
     @override
     def convert_stream_full_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, Any, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str, Any]:
         """
         Convert stream full response.
         :param stream_response: stream response
@@ -112,8 +112,8 @@ class AdvancedChatAppGenerateResponseConverter(
     @classmethod
     @override
     def convert_stream_simple_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, Any, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str, Any]:
         """
         Convert stream simple response.
         :param stream_response: stream response

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

@@ -222,7 +222,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
     ) -> Union[
         ChatbotAppBlockingResponse,
         AdvancedChatPausedBlockingResponse,
-        Generator[ChatbotAppStreamResponse, None, None],
+        Generator[ChatbotAppStreamResponse],
     ]:
         """
         Process generate task pipeline.
@@ -242,7 +242,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
             return self._to_blocking_response(generator)
 
     def _to_blocking_response(
-        self, generator: Generator[StreamResponse, None, None]
+        self, generator: Generator[StreamResponse]
     ) -> Union[ChatbotAppBlockingResponse, AdvancedChatPausedBlockingResponse]:
         """
         Process blocking response.
@@ -330,9 +330,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
             ),
         )
 
-    def _to_stream_response(
-        self, generator: Generator[StreamResponse, None, None]
-    ) -> Generator[ChatbotAppStreamResponse, Any, None]:
+    def _to_stream_response(self, generator: Generator[StreamResponse]) -> Generator[ChatbotAppStreamResponse, Any]:
         """
         To stream response.
         :return:
@@ -359,7 +357,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
 
     def _wrapper_process_stream_response(
         self, trace_manager: TraceQueueManager | None = None
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         tts_publisher = None
         task_id = self._application_generate_entity.task_id
         tenant_id = self._application_generate_entity.app_config.tenant_id
@@ -430,11 +428,11 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         if not self._workflow_run_id:
             raise ValueError("workflow run not initialized.")
 
-    def _handle_ping_event(self, event: QueuePingEvent, **kwargs) -> Generator[PingStreamResponse, None, None]:
+    def _handle_ping_event(self, event: QueuePingEvent, **kwargs) -> Generator[PingStreamResponse]:
         """Handle ping events."""
         yield self._base_task_pipeline.ping_stream_response()
 
-    def _handle_error_event(self, event: QueueErrorEvent, **kwargs) -> Generator[ErrorStreamResponse, None, None]:
+    def _handle_error_event(self, event: QueueErrorEvent, **kwargs) -> Generator[ErrorStreamResponse]:
         """Handle error events."""
         with self._database_session() as session:
             err = self._base_task_pipeline.handle_error(event=event, session=session, message_id=self._message_id)
@@ -444,7 +442,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         self,
         event: QueueWorkflowStartedEvent,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle workflow started events."""
         runtime_state = self._resolve_graph_runtime_state()
         run_id = self._extract_workflow_run_id(runtime_state)
@@ -462,7 +460,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
 
         yield workflow_start_resp
 
-    def _handle_node_retry_event(self, event: QueueNodeRetryEvent, **kwargs) -> Generator[StreamResponse, None, None]:
+    def _handle_node_retry_event(self, event: QueueNodeRetryEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle node retry events."""
         self._ensure_workflow_initialized()
 
@@ -474,9 +472,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         if node_retry_resp:
             yield node_retry_resp
 
-    def _handle_node_started_event(
-        self, event: QueueNodeStartedEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_node_started_event(self, event: QueueNodeStartedEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle node started events."""
         self._ensure_workflow_initialized()
 
@@ -488,9 +484,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         if node_start_resp:
             yield node_start_resp
 
-    def _handle_node_succeeded_event(
-        self, event: QueueNodeSucceededEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_node_succeeded_event(self, event: QueueNodeSucceededEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle node succeeded events."""
         # Record files if it's an answer node or end node
         if event.node_type in [BuiltinNodeTypes.ANSWER, BuiltinNodeTypes.END, BuiltinNodeTypes.LLM]:
@@ -523,7 +517,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         self,
         event: Union[QueueNodeFailedEvent, QueueNodeExceptionEvent],
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle various node failure events."""
         node_finish_resp = self._workflow_response_converter.workflow_node_finish_to_stream_response(
             event=event,
@@ -543,7 +537,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         tts_publisher: AppGeneratorTTSPublisher | None = None,
         queue_message: Union[WorkflowQueueMessage, MessageQueueMessage] | None = None,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle text chunk events."""
         delta_text = event.text
         if delta_text is None:
@@ -571,9 +565,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
             answer=delta_text, message_id=self._message_id, from_variable_selector=event.from_variable_selector
         )
 
-    def _handle_reasoning_chunk_event(
-        self, event: QueueReasoningChunkEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_reasoning_chunk_event(self, event: QueueReasoningChunkEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle out-of-band reasoning chunk events.
 
         Pure emit: reasoning is streamed on its own channel and never written to the
@@ -592,9 +584,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
             ),
         )
 
-    def _handle_iteration_start_event(
-        self, event: QueueIterationStartEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_iteration_start_event(self, event: QueueIterationStartEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle iteration start events."""
         self._ensure_workflow_initialized()
 
@@ -605,9 +595,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         )
         yield iter_start_resp
 
-    def _handle_iteration_next_event(
-        self, event: QueueIterationNextEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_iteration_next_event(self, event: QueueIterationNextEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle iteration next events."""
         self._ensure_workflow_initialized()
 
@@ -620,7 +608,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
 
     def _handle_iteration_completed_event(
         self, event: QueueIterationCompletedEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle iteration completed events."""
         self._ensure_workflow_initialized()
 
@@ -631,7 +619,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         )
         yield iter_finish_resp
 
-    def _handle_loop_start_event(self, event: QueueLoopStartEvent, **kwargs) -> Generator[StreamResponse, None, None]:
+    def _handle_loop_start_event(self, event: QueueLoopStartEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle loop start events."""
         self._ensure_workflow_initialized()
 
@@ -642,7 +630,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         )
         yield loop_start_resp
 
-    def _handle_loop_next_event(self, event: QueueLoopNextEvent, **kwargs) -> Generator[StreamResponse, None, None]:
+    def _handle_loop_next_event(self, event: QueueLoopNextEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle loop next events."""
         self._ensure_workflow_initialized()
 
@@ -653,9 +641,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         )
         yield loop_next_resp
 
-    def _handle_loop_completed_event(
-        self, event: QueueLoopCompletedEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_loop_completed_event(self, event: QueueLoopCompletedEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle loop completed events."""
         self._ensure_workflow_initialized()
 
@@ -672,7 +658,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         *,
         trace_manager: TraceQueueManager | None = None,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle workflow succeeded events."""
         _ = trace_manager
         self._ensure_workflow_initialized()
@@ -695,7 +681,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         *,
         trace_manager: TraceQueueManager | None = None,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle workflow partial success events."""
         _ = trace_manager
         self._ensure_workflow_initialized()
@@ -717,7 +703,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         self,
         event: QueueWorkflowPausedEvent,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle workflow paused events."""
         validated_state = self._ensure_graph_runtime_initialized()
         responses = self._workflow_response_converter.workflow_pause_to_stream_response(
@@ -749,7 +735,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         *,
         trace_manager: TraceQueueManager | None = None,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle workflow failed events."""
         _ = trace_manager
         self._ensure_workflow_initialized()
@@ -778,7 +764,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         graph_runtime_state: GraphRuntimeState | None = None,
         trace_manager: TraceQueueManager | None = None,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle stop events."""
         _ = trace_manager
         resolved_state = None
@@ -818,7 +804,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         *,
         graph_runtime_state: GraphRuntimeState | None = None,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle advanced chat message end events."""
         resolved_state = self._ensure_graph_runtime_initialized(graph_runtime_state)
 
@@ -842,28 +828,24 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
 
     def _handle_retriever_resources_event(
         self, event: QueueRetrieverResourcesEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle retriever resources events."""
         self._message_cycle_manager.handle_retriever_resources(event)
         yield from ()
 
-    def _handle_annotation_reply_event(
-        self, event: QueueAnnotationReplyEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_annotation_reply_event(self, event: QueueAnnotationReplyEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle annotation reply events."""
         with self._database_session() as session:
             self._message_cycle_manager.handle_annotation_reply(event, session)
         yield from ()
 
-    def _handle_message_replace_event(
-        self, event: QueueMessageReplaceEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_message_replace_event(self, event: QueueMessageReplaceEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle message replace events."""
         yield self._message_cycle_manager.message_replace_to_stream_response(answer=event.text, reason=event.reason)
 
     def _handle_human_input_form_filled_event(
         self, event: QueueHumanInputFormFilledEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle human input form filled events."""
         self._persist_human_input_extra_content(node_id=event.node_id)
         yield self._workflow_response_converter.human_input_form_filled_to_stream_response(
@@ -872,7 +854,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
 
     def _handle_human_input_form_timeout_event(
         self, event: QueueHumanInputFormTimeoutEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle human input form timeout events."""
         yield self._workflow_response_converter.human_input_form_timeout_to_stream_response(
             event=event, task_id=self._application_generate_entity.task_id
@@ -920,7 +902,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
             return None
         return form.id
 
-    def _handle_agent_log_event(self, event: QueueAgentLogEvent, **kwargs) -> Generator[StreamResponse, None, None]:
+    def _handle_agent_log_event(self, event: QueueAgentLogEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle agent log events."""
         yield self._workflow_response_converter.handle_agent_log(
             task_id=self._application_generate_entity.task_id, event=event
@@ -971,7 +953,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         tts_publisher: AppGeneratorTTSPublisher | None = None,
         trace_manager: TraceQueueManager | None = None,
         queue_message: Union[WorkflowQueueMessage, MessageQueueMessage] | None = None,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Dispatch events using elegant pattern matching."""
         handlers = self._get_event_handlers()
         event_type = type(event)
@@ -1009,7 +991,7 @@ class AdvancedChatAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         self,
         tts_publisher: AppGeneratorTTSPublisher | None = None,
         trace_manager: TraceQueueManager | None = None,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """
         Process stream response using elegant Fluent Python patterns.
         Maintains exact same functionality as original 57-if-statement version.

api/core/app/apps/agent_app/app_generator.py

@@ -85,7 +85,7 @@ class AgentAppGenerator(MessageBasedAppGenerator):
         invoke_from: InvokeFrom,
         session: Session,
         streaming: bool = True,
-    ) -> Mapping[str, Any] | Generator[Mapping | str, None, None]:
+    ) -> Mapping[str, Any] | Generator[Mapping | str]:
         if not streaming:
             raise AgentAppGeneratorError("Agent App only supports streaming mode")
 

api/core/app/apps/agent_chat/app_generator.py

@@ -57,7 +57,7 @@ class AgentChatAppGenerator(MessageBasedAppGenerator):
         invoke_from: InvokeFrom,
         streaming: Literal[True],
         session: Session,
-    ) -> Generator[Mapping | str, None, None]: ...
+    ) -> Generator[Mapping | str]: ...
 
     @overload
     def generate(
@@ -69,7 +69,7 @@ class AgentChatAppGenerator(MessageBasedAppGenerator):
         invoke_from: InvokeFrom,
         streaming: bool,
         session: Session,
-    ) -> Mapping | Generator[Mapping | str, None, None]: ...
+    ) -> Mapping | Generator[Mapping | str]: ...
 
     def generate(
         self,
@@ -80,7 +80,7 @@ class AgentChatAppGenerator(MessageBasedAppGenerator):
         invoke_from: InvokeFrom,
         streaming: bool = True,
         session: Session,
-    ) -> Mapping | Generator[Mapping | str, None, None]:
+    ) -> Mapping | Generator[Mapping | str]:
         """
         Generate App response.
 

api/core/app/apps/agent_chat/generate_response_converter.py

@@ -58,8 +58,8 @@ class AgentChatAppGenerateResponseConverter(AppGenerateResponseConverter[Chatbot
     @classmethod
     @override
     def convert_stream_full_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, None, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str]:
         """
         Convert stream full response.
         :param stream_response: stream response
@@ -90,8 +90,8 @@ class AgentChatAppGenerateResponseConverter(AppGenerateResponseConverter[Chatbot
     @classmethod
     @override
     def convert_stream_simple_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, None, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str]:
         """
         Convert stream simple response.
         :param stream_response: stream response

api/core/app/apps/base_app_generate_response_converter.py

@@ -23,14 +23,14 @@ class AppGenerateResponseConverter[TBlockingResponse: AppBlockingResponse](ABC):
 
     @classmethod
     def convert(
-        cls, response: Union[AppBlockingResponse, Generator[AppStreamResponse, Any, None]], invoke_from: InvokeFrom
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], Any, None]:
+        cls, response: Union[AppBlockingResponse, Generator[AppStreamResponse, Any]], invoke_from: InvokeFrom
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], Any]:
         if invoke_from in {InvokeFrom.DEBUGGER, InvokeFrom.SERVICE_API, InvokeFrom.OPENAPI}:
             if isinstance(response, AppBlockingResponse):
                 return cls.convert_blocking_full_response(cls._cast_blocking_response(response))
             else:
 
-                def _generate_full_response() -> Generator[dict[str, Any] | str, Any, None]:
+                def _generate_full_response() -> Generator[dict[str, Any] | str, Any]:
                     yield from cls.convert_stream_full_response(response)
 
                 return _generate_full_response()
@@ -39,7 +39,7 @@ class AppGenerateResponseConverter[TBlockingResponse: AppBlockingResponse](ABC):
                 return cls.convert_blocking_simple_response(cls._cast_blocking_response(response))
             else:
 
-                def _generate_simple_response() -> Generator[dict[str, Any] | str, Any, None]:
+                def _generate_simple_response() -> Generator[dict[str, Any] | str, Any]:
                     yield from cls.convert_stream_simple_response(response)
 
                 return _generate_simple_response()
@@ -57,15 +57,15 @@ class AppGenerateResponseConverter[TBlockingResponse: AppBlockingResponse](ABC):
     @classmethod
     @abstractmethod
     def convert_stream_full_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, None, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str]:
         raise NotImplementedError
 
     @classmethod
     @abstractmethod
     def convert_stream_simple_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, None, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str]:
         raise NotImplementedError
 
     @classmethod

api/core/app/apps/base_app_generator.py

@@ -94,9 +94,9 @@ class BaseAppGenerator:
 
     @staticmethod
     def _wrap_stream_with_worker_thread_join[ResponseT](
-        response_stream: Generator[ResponseT, None, None],
+        response_stream: Generator[ResponseT],
         worker_thread: threading.Thread,
-    ) -> Generator[ResponseT, None, None]:
+    ) -> Generator[ResponseT]:
         """Keep the producer owned by the response stream until both finish."""
         try:
             yield from response_stream
@@ -310,7 +310,7 @@ class BaseAppGenerator:
         return value
 
     @classmethod
-    def convert_to_event_stream(cls, generator: Union[Mapping, Generator[Mapping | str, None, None]]):
+    def convert_to_event_stream(cls, generator: Union[Mapping, Generator[Mapping | str]]):
         """
         Convert messages into event stream
         """

api/core/app/apps/base_app_runner.py

@@ -214,7 +214,7 @@ class AppRunner:
 
     def _handle_invoke_result(
         self,
-        invoke_result: Union[LLMResult, Generator[Any, None, None]],
+        invoke_result: Union[LLMResult, Generator[Any]],
         queue_manager: AppQueueManager,
         stream: bool,
         agent: bool = False,
@@ -275,7 +275,7 @@ class AppRunner:
 
     def _handle_invoke_result_stream(
         self,
-        invoke_result: Generator[LLMResultChunk, None, None],
+        invoke_result: Generator[LLMResultChunk],
         queue_manager: AppQueueManager,
         agent: bool,
         message_id: str | None = None,

api/core/app/apps/chat/app_generator.py

@@ -45,7 +45,7 @@ class ChatAppGenerator(MessageBasedAppGenerator):
         streaming: Literal[True],
         *,
         session: Session,
-    ) -> Generator[Mapping | str, None, None]: ...
+    ) -> Generator[Mapping | str]: ...
 
     @overload
     def generate(
@@ -69,7 +69,7 @@ class ChatAppGenerator(MessageBasedAppGenerator):
         streaming: bool,
         *,
         session: Session,
-    ) -> Mapping[str, Any] | Generator[Mapping[str, Any] | str, None, None]: ...
+    ) -> Mapping[str, Any] | Generator[Mapping[str, Any] | str]: ...
 
     def generate(
         self,
@@ -80,7 +80,7 @@ class ChatAppGenerator(MessageBasedAppGenerator):
         streaming: bool = True,
         *,
         session: Session,
-    ) -> Mapping[str, Any] | Generator[Mapping[str, Any] | str, None, None]:
+    ) -> Mapping[str, Any] | Generator[Mapping[str, Any] | str]:
         """
         Generate App response.
 

api/core/app/apps/chat/generate_response_converter.py

@@ -58,8 +58,8 @@ class ChatAppGenerateResponseConverter(AppGenerateResponseConverter[ChatbotAppBl
     @classmethod
     @override
     def convert_stream_full_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, None, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str]:
         """
         Convert stream full response.
         :param stream_response: stream response
@@ -90,8 +90,8 @@ class ChatAppGenerateResponseConverter(AppGenerateResponseConverter[ChatbotAppBl
     @classmethod
     @override
     def convert_stream_simple_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, None, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str]:
         """
         Convert stream simple response.
         :param stream_response: stream response

api/core/app/apps/completion/app_generator.py

@@ -46,7 +46,7 @@ class CompletionAppGenerator(MessageBasedAppGenerator):
         streaming: Literal[True],
         *,
         session: Session,
-    ) -> Generator[str | Mapping[str, Any], None, None]: ...
+    ) -> Generator[str | Mapping[str, Any]]: ...
 
     @overload
     def generate(
@@ -70,7 +70,7 @@ class CompletionAppGenerator(MessageBasedAppGenerator):
         streaming: bool = False,
         *,
         session: Session,
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], None, None]: ...
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any]]: ...
 
     def generate(
         self,
@@ -81,7 +81,7 @@ class CompletionAppGenerator(MessageBasedAppGenerator):
         streaming: bool = True,
         *,
         session: Session,
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], None, None]:
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any]]:
         """
         Generate App response.
 
@@ -280,7 +280,7 @@ class CompletionAppGenerator(MessageBasedAppGenerator):
         stream: bool = True,
         *,
         session: Session,
-    ) -> Mapping | Generator[Mapping | str, None, None]:
+    ) -> Mapping | Generator[Mapping | str]:
         """
         Generate App response.
 

api/core/app/apps/completion/generate_response_converter.py

@@ -57,8 +57,8 @@ class CompletionAppGenerateResponseConverter(AppGenerateResponseConverter[Comple
     @classmethod
     @override
     def convert_stream_full_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, None, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str]:
         """
         Convert stream full response.
         :param stream_response: stream response
@@ -88,8 +88,8 @@ class CompletionAppGenerateResponseConverter(AppGenerateResponseConverter[Comple
     @classmethod
     @override
     def convert_stream_simple_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, None, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str]:
         """
         Convert stream simple response.
         :param stream_response: stream response

api/core/app/apps/message_based_app_generator.py

@@ -59,7 +59,7 @@ class MessageBasedAppGenerator(BaseAppGenerator):
     ) -> Union[
         ChatbotAppBlockingResponse,
         CompletionAppBlockingResponse,
-        Generator[Union[ChatbotAppStreamResponse, CompletionAppStreamResponse], None, None],
+        Generator[Union[ChatbotAppStreamResponse, CompletionAppStreamResponse]],
     ]:
         """
         Handle response.
@@ -321,7 +321,7 @@ class MessageBasedAppGenerator(BaseAppGenerator):
         workflow_run_id: str,
         idle_timeout=300,
         on_subscribe: Callable[[], None] | None = None,
-    ) -> Generator[Mapping | str, None, None]:
+    ) -> Generator[Mapping | str]:
         topic = cls.get_response_topic(app_mode, workflow_run_id)
         subscriber = topic.as_subscriber()
         subscription = (

api/core/app/apps/message_generator.py

@@ -28,7 +28,7 @@ class MessageGenerator:
         ping_interval: float = 10.0,
         on_subscribe: Callable[[], None] | None = None,
         terminal_events: Iterable[str | StreamEvent] | None = None,
-    ) -> Generator[Mapping | str, None, None]:
+    ) -> Generator[Mapping | str]:
         topic = cls.get_response_topic(app_mode, workflow_run_id)
         subscriber = topic.as_subscriber()
         subscription = (

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

@@ -37,8 +37,8 @@ class WorkflowAppGenerateResponseConverter(AppGenerateResponseConverter[Workflow
     @classmethod
     @override
     def convert_stream_full_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, None, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str]:
         """
         Convert stream full response.
         :param stream_response: stream response
@@ -71,8 +71,8 @@ class WorkflowAppGenerateResponseConverter(AppGenerateResponseConverter[Workflow
     @classmethod
     @override
     def convert_stream_simple_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, None, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str]:
         """
         Convert stream simple response.
         :param stream_response: stream response

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

@@ -78,7 +78,7 @@ class PipelineGenerator(BaseAppGenerator):
         call_depth: int,
         workflow_thread_pool_id: str | None,
         is_retry: bool = False,
-    ) -> Generator[Mapping | str, None, None]: ...
+    ) -> Generator[Mapping | str]: ...
 
     @overload
     def generate(
@@ -110,7 +110,7 @@ class PipelineGenerator(BaseAppGenerator):
         call_depth: int,
         workflow_thread_pool_id: str | None,
         is_retry: bool = False,
-    ) -> Mapping[str, Any] | Generator[Mapping | str, None, None]: ...
+    ) -> Mapping[str, Any] | Generator[Mapping | str]: ...
 
     def generate(
         self,
@@ -125,7 +125,7 @@ class PipelineGenerator(BaseAppGenerator):
         call_depth: int = 0,
         workflow_thread_pool_id: str | None = None,
         is_retry: bool = False,
-    ) -> Mapping[str, Any] | Generator[Mapping | str, None, None] | None:
+    ) -> Mapping[str, Any] | Generator[Mapping | str] | None:
         # Add null check for dataset
 
         dataset = self._documents.get_pipeline_dataset(pipeline, session=session)
@@ -306,7 +306,7 @@ class PipelineGenerator(BaseAppGenerator):
         streaming: bool = True,
         variable_loader: VariableLoader = DUMMY_VARIABLE_LOADER,
         workflow_thread_pool_id: str | None = None,
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], None, None]:
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any]]:
         """
         Generate App response.
 
@@ -388,7 +388,7 @@ class PipelineGenerator(BaseAppGenerator):
         streaming: bool = True,
         *,
         session: Session,
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], None, None]:
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any]]:
         """
         Generate App response.
 
@@ -490,7 +490,7 @@ class PipelineGenerator(BaseAppGenerator):
         streaming: bool = True,
         *,
         session: Session,
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], None, None]:
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any]]:
         """
         Generate App response.
 
@@ -668,11 +668,7 @@ class PipelineGenerator(BaseAppGenerator):
         user: Account | EndUser,
         draft_var_saver_factory: DraftVariableSaverFactory,
         stream: bool = False,
-    ) -> (
-        WorkflowAppBlockingResponse
-        | WorkflowAppPausedBlockingResponse
-        | Generator[WorkflowAppStreamResponse, None, None]
-    ):
+    ) -> WorkflowAppBlockingResponse | WorkflowAppPausedBlockingResponse | Generator[WorkflowAppStreamResponse]:
         """
         Handle response.
         :param application_generate_entity: application generate entity

api/core/app/apps/streaming_utils.py

@@ -17,7 +17,7 @@ def stream_topic_events(
     ping_interval: float | None = None,
     on_subscribe: Callable[[], None] | None = None,
     terminal_events: Iterable[str | StreamEvent] | None = None,
-) -> Generator[Mapping[str, Any] | str, None, None]:
+) -> Generator[Mapping[str, Any] | str]:
     # send a PING event immediately to prevent the connection staying in pending state for a long time.
     #
     # This simplify the debugging process as the DevTools in Chrome does not

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

@@ -105,7 +105,7 @@ class WorkflowAppGenerator(BaseAppGenerator):
         root_node_id: str | None = None,
         graph_engine_layers: Sequence[GraphEngineLayer] = (),
         pause_state_config: PauseStateLayerConfig | None = None,
-    ) -> Generator[Mapping[str, Any] | str, None, None]: ...
+    ) -> Generator[Mapping[str, Any] | str]: ...
 
     @overload
     def generate(
@@ -141,7 +141,7 @@ class WorkflowAppGenerator(BaseAppGenerator):
         root_node_id: str | None = None,
         graph_engine_layers: Sequence[GraphEngineLayer] = (),
         pause_state_config: PauseStateLayerConfig | None = None,
-    ) -> Mapping[str, Any] | Generator[Mapping[str, Any] | str, None, None]: ...
+    ) -> Mapping[str, Any] | Generator[Mapping[str, Any] | str]: ...
 
     def generate(
         self,
@@ -158,7 +158,7 @@ class WorkflowAppGenerator(BaseAppGenerator):
         root_node_id: str | None = None,
         graph_engine_layers: Sequence[GraphEngineLayer] = (),
         pause_state_config: PauseStateLayerConfig | None = None,
-    ) -> Mapping[str, Any] | Generator[Mapping[str, Any] | str, None, None]:
+    ) -> Mapping[str, Any] | Generator[Mapping[str, Any] | str]:
         with self._bind_file_access_scope(tenant_id=app_model.tenant_id, user=user, invoke_from=invoke_from):
             files: Sequence[Mapping[str, Any]] = args.get("files") or []
 
@@ -283,7 +283,7 @@ class WorkflowAppGenerator(BaseAppGenerator):
         pause_state_config: PauseStateLayerConfig | None = None,
         variable_loader: VariableLoader = DUMMY_VARIABLE_LOADER,
         response_stream_filter: ResponseStreamFilter | None = None,
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], None, None]:
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any]]:
         """
         Resume a paused workflow execution using the persisted runtime state.
 
@@ -333,7 +333,7 @@ class WorkflowAppGenerator(BaseAppGenerator):
         graph_runtime_state: GraphRuntimeState | None = None,
         pause_state_config: PauseStateLayerConfig | None = None,
         response_stream_filter: ResponseStreamFilter | None = None,
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], None, None]:
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any]]:
         """
         Generate App response.
 
@@ -436,7 +436,7 @@ class WorkflowAppGenerator(BaseAppGenerator):
         streaming: bool = True,
         *,
         session: Session,
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], None, None]:
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any]]:
         """
         Generate App response.
 
@@ -530,7 +530,7 @@ class WorkflowAppGenerator(BaseAppGenerator):
         streaming: bool = True,
         *,
         session: Session,
-    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any], None, None]:
+    ) -> Mapping[str, Any] | Generator[str | Mapping[str, Any]]:
         """
         Generate App response.
 
@@ -710,11 +710,7 @@ class WorkflowAppGenerator(BaseAppGenerator):
         user: Account | EndUser,
         draft_var_saver_factory: DraftVariableSaverFactory,
         stream: bool = False,
-    ) -> (
-        WorkflowAppBlockingResponse
-        | WorkflowAppPausedBlockingResponse
-        | Generator[WorkflowAppStreamResponse, None, None]
-    ):
+    ) -> WorkflowAppBlockingResponse | WorkflowAppPausedBlockingResponse | Generator[WorkflowAppStreamResponse]:
         """
         Handle response.
         :param application_generate_entity: application generate entity

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

@@ -44,8 +44,8 @@ class WorkflowAppGenerateResponseConverter(
     @classmethod
     @override
     def convert_stream_full_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, None, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str]:
         """
         Convert stream full response.
         :param stream_response: stream response
@@ -78,8 +78,8 @@ class WorkflowAppGenerateResponseConverter(
     @classmethod
     @override
     def convert_stream_simple_response(
-        cls, stream_response: Generator[AppStreamResponse, None, None]
-    ) -> Generator[dict[str, Any] | str, None, None]:
+        cls, stream_response: Generator[AppStreamResponse]
+    ) -> Generator[dict[str, Any] | str]:
         """
         Convert stream simple response.
         :param stream_response: stream response

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

@@ -124,9 +124,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
 
     def process(
         self,
-    ) -> Union[
-        WorkflowAppBlockingResponse, WorkflowAppPausedBlockingResponse, Generator[WorkflowAppStreamResponse, None, None]
-    ]:
+    ) -> Union[WorkflowAppBlockingResponse, WorkflowAppPausedBlockingResponse, Generator[WorkflowAppStreamResponse]]:
         """
         Process generate task pipeline.
         :return:
@@ -138,7 +136,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
             return self._to_blocking_response(generator)
 
     def _to_blocking_response(
-        self, generator: Generator[StreamResponse, None, None]
+        self, generator: Generator[StreamResponse]
     ) -> Union[WorkflowAppBlockingResponse, WorkflowAppPausedBlockingResponse]:
         """
         To blocking response.
@@ -227,9 +225,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
             ),
         )
 
-    def _to_stream_response(
-        self, generator: Generator[StreamResponse, None, None]
-    ) -> Generator[WorkflowAppStreamResponse, None, None]:
+    def _to_stream_response(self, generator: Generator[StreamResponse]) -> Generator[WorkflowAppStreamResponse]:
         """
         To stream response.
         :return:
@@ -255,7 +251,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
 
     def _wrapper_process_stream_response(
         self, trace_manager: TraceQueueManager | None = None
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         tts_publisher = None
         task_id = self._application_generate_entity.task_id
         tenant_id = self._application_generate_entity.app_config.tenant_id
@@ -321,18 +317,16 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         if not self._workflow_execution_id:
             raise ValueError("workflow run not initialized.")
 
-    def _handle_ping_event(self, event: QueuePingEvent, **kwargs) -> Generator[PingStreamResponse, None, None]:
+    def _handle_ping_event(self, event: QueuePingEvent, **kwargs) -> Generator[PingStreamResponse]:
         """Handle ping events."""
         yield self._base_task_pipeline.ping_stream_response()
 
-    def _handle_error_event(self, event: QueueErrorEvent, **kwargs) -> Generator[ErrorStreamResponse, None, None]:
+    def _handle_error_event(self, event: QueueErrorEvent, **kwargs) -> Generator[ErrorStreamResponse]:
         """Handle error events."""
         err = self._base_task_pipeline.handle_error(event=event)
         yield self._base_task_pipeline.error_to_stream_response(err)
 
-    def _handle_workflow_started_event(
-        self, event: QueueWorkflowStartedEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_workflow_started_event(self, event: QueueWorkflowStartedEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle workflow started events."""
         runtime_state = self._resolve_graph_runtime_state()
 
@@ -351,7 +345,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         )
         yield start_resp
 
-    def _handle_node_retry_event(self, event: QueueNodeRetryEvent, **kwargs) -> Generator[StreamResponse, None, None]:
+    def _handle_node_retry_event(self, event: QueueNodeRetryEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle node retry events."""
         self._ensure_workflow_initialized()
 
@@ -363,9 +357,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         if response:
             yield response
 
-    def _handle_node_started_event(
-        self, event: QueueNodeStartedEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_node_started_event(self, event: QueueNodeStartedEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle node started events."""
         self._ensure_workflow_initialized()
 
@@ -377,9 +369,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         if node_start_response:
             yield node_start_response
 
-    def _handle_node_succeeded_event(
-        self, event: QueueNodeSucceededEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_node_succeeded_event(self, event: QueueNodeSucceededEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle node succeeded events."""
         node_success_response = self._workflow_response_converter.workflow_node_finish_to_stream_response(
             event=event,
@@ -395,7 +385,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         self,
         event: Union[QueueNodeFailedEvent, QueueNodeExceptionEvent],
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle various node failure events."""
         node_failed_response = self._workflow_response_converter.workflow_node_finish_to_stream_response(
             event=event,
@@ -408,9 +398,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         if node_failed_response:
             yield node_failed_response
 
-    def _handle_iteration_start_event(
-        self, event: QueueIterationStartEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_iteration_start_event(self, event: QueueIterationStartEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle iteration start events."""
         self._ensure_workflow_initialized()
 
@@ -421,9 +409,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         )
         yield iter_start_resp
 
-    def _handle_iteration_next_event(
-        self, event: QueueIterationNextEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_iteration_next_event(self, event: QueueIterationNextEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle iteration next events."""
         self._ensure_workflow_initialized()
 
@@ -436,7 +422,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
 
     def _handle_iteration_completed_event(
         self, event: QueueIterationCompletedEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle iteration completed events."""
         self._ensure_workflow_initialized()
 
@@ -447,7 +433,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         )
         yield iter_finish_resp
 
-    def _handle_loop_start_event(self, event: QueueLoopStartEvent, **kwargs) -> Generator[StreamResponse, None, None]:
+    def _handle_loop_start_event(self, event: QueueLoopStartEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle loop start events."""
         self._ensure_workflow_initialized()
 
@@ -458,7 +444,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         )
         yield loop_start_resp
 
-    def _handle_loop_next_event(self, event: QueueLoopNextEvent, **kwargs) -> Generator[StreamResponse, None, None]:
+    def _handle_loop_next_event(self, event: QueueLoopNextEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle loop next events."""
         self._ensure_workflow_initialized()
 
@@ -469,9 +455,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         )
         yield loop_next_resp
 
-    def _handle_loop_completed_event(
-        self, event: QueueLoopCompletedEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_loop_completed_event(self, event: QueueLoopCompletedEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle loop completed events."""
         self._ensure_workflow_initialized()
 
@@ -488,7 +472,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         *,
         trace_manager: TraceQueueManager | None = None,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle workflow succeeded events."""
         _ = trace_manager
         self._ensure_workflow_initialized()
@@ -508,7 +492,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         *,
         trace_manager: TraceQueueManager | None = None,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle workflow partial success events."""
         _ = trace_manager
         self._ensure_workflow_initialized()
@@ -526,7 +510,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         self,
         event: QueueWorkflowPausedEvent,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle workflow paused events."""
         self._ensure_workflow_initialized()
         validated_state = self._ensure_graph_runtime_initialized()
@@ -543,7 +527,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         *,
         trace_manager: TraceQueueManager | None = None,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle workflow failed and stop events."""
         _ = trace_manager
         self._ensure_workflow_initialized()
@@ -574,7 +558,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         tts_publisher: AppGeneratorTTSPublisher | None = None,
         queue_message: Union[WorkflowQueueMessage, MessageQueueMessage] | None = None,
         **kwargs,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle text chunk events."""
         delta_text = event.text
         if delta_text is None:
@@ -586,9 +570,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
 
         yield self._text_chunk_to_stream_response(delta_text, from_variable_selector=event.from_variable_selector)
 
-    def _handle_reasoning_chunk_event(
-        self, event: QueueReasoningChunkEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    def _handle_reasoning_chunk_event(self, event: QueueReasoningChunkEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle reasoning chunk events."""
         # is_final with empty reasoning is still forwarded as the "thinking finished" signal
         if not event.reasoning and not event.is_final:
@@ -602,7 +584,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
             ),
         )
 
-    def _handle_agent_log_event(self, event: QueueAgentLogEvent, **kwargs) -> Generator[StreamResponse, None, None]:
+    def _handle_agent_log_event(self, event: QueueAgentLogEvent, **kwargs) -> Generator[StreamResponse]:
         """Handle agent log events."""
         yield self._workflow_response_converter.handle_agent_log(
             task_id=self._application_generate_entity.task_id, event=event
@@ -610,7 +592,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
 
     def _handle_human_input_form_filled_event(
         self, event: QueueHumanInputFormFilledEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle human input form filled events."""
         yield self._workflow_response_converter.human_input_form_filled_to_stream_response(
             event=event, task_id=self._application_generate_entity.task_id
@@ -618,7 +600,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
 
     def _handle_human_input_form_timeout_event(
         self, event: QueueHumanInputFormTimeoutEvent, **kwargs
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Handle human input form timeout events."""
         yield self._workflow_response_converter.human_input_form_timeout_to_stream_response(
             event=event, task_id=self._application_generate_entity.task_id
@@ -662,7 +644,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         tts_publisher: AppGeneratorTTSPublisher | None = None,
         trace_manager: TraceQueueManager | None = None,
         queue_message: Union[WorkflowQueueMessage, MessageQueueMessage] | None = None,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """Dispatch events using elegant pattern matching."""
         handlers = self._get_event_handlers()
         event_type = type(event)
@@ -710,7 +692,7 @@ class WorkflowAppGenerateTaskPipeline(GraphRuntimeStateSupport):
         self,
         tts_publisher: AppGeneratorTTSPublisher | None = None,
         trace_manager: TraceQueueManager | None = None,
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """
         Process stream response using elegant Fluent Python patterns.
         Maintains exact same functionality as original 44-if-statement version.

api/core/app/features/rate_limiting/rate_limit.py

@@ -99,7 +99,7 @@ class RateLimit:
     def gen_request_key() -> str:
         return str(uuid.uuid4())
 
-    def generate(self, generator: Union[Generator[str, None, None], Mapping[str, Any]], request_id: str):
+    def generate(self, generator: Union[Generator[str], Mapping[str, Any]], request_id: str):
         if isinstance(generator, Mapping):
             return generator
         else:
@@ -119,7 +119,7 @@ def rate_limit_context(rate_limit: RateLimit, request_id: str | None):
 
 
 class RateLimitGenerator:
-    def __init__(self, rate_limit: RateLimit, generator: Generator[str, None, None], request_id: str):
+    def __init__(self, rate_limit: RateLimit, generator: Generator[str], request_id: str):
         self.rate_limit = rate_limit
         self.generator = generator
         self.request_id = request_id

api/core/app/file_access/scope.py

@@ -108,7 +108,7 @@ def is_retriever_segment_access_granted(segment_id: str) -> bool:
 
 
 @contextmanager
-def bind_file_access_scope(scope: FileAccessScope) -> Generator[None, None, None]:
+def bind_file_access_scope(scope: FileAccessScope) -> Generator[None]:
     token = _current_file_access_scope.set(scope)
     try:
         yield

api/core/app/task_pipeline/easy_ui_based_generate_task_pipeline.py

@@ -120,7 +120,7 @@ class EasyUIBasedGenerateTaskPipeline(BasedGenerateTaskPipeline[EasyUIAppGenerat
     ) -> (
         ChatbotAppBlockingResponse
         | CompletionAppBlockingResponse
-        | Generator[ChatbotAppStreamResponse | CompletionAppStreamResponse, None, None]
+        | Generator[ChatbotAppStreamResponse | CompletionAppStreamResponse]
     ):
         if self._application_generate_entity.app_config.app_mode != AppMode.COMPLETION:
             # start generate conversation name thread
@@ -137,7 +137,7 @@ class EasyUIBasedGenerateTaskPipeline(BasedGenerateTaskPipeline[EasyUIAppGenerat
             return self._to_blocking_response(generator)
 
     def _to_blocking_response(
-        self, generator: Generator[StreamResponse, None, None]
+        self, generator: Generator[StreamResponse]
     ) -> ChatbotAppBlockingResponse | CompletionAppBlockingResponse:
         """
         Process blocking response.
@@ -185,8 +185,8 @@ class EasyUIBasedGenerateTaskPipeline(BasedGenerateTaskPipeline[EasyUIAppGenerat
         raise RuntimeError("queue listening stopped unexpectedly.")
 
     def _to_stream_response(
-        self, generator: Generator[StreamResponse, None, None]
-    ) -> Generator[ChatbotAppStreamResponse | CompletionAppStreamResponse, None, None]:
+        self, generator: Generator[StreamResponse]
+    ) -> Generator[ChatbotAppStreamResponse | CompletionAppStreamResponse]:
         """
         To stream response.
         :return:
@@ -220,7 +220,7 @@ class EasyUIBasedGenerateTaskPipeline(BasedGenerateTaskPipeline[EasyUIAppGenerat
 
     def _wrapper_process_stream_response(
         self, trace_manager: TraceQueueManager | None = None
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         tenant_id = self._application_generate_entity.app_config.tenant_id
         task_id = self._application_generate_entity.task_id
         publisher = None
@@ -273,7 +273,7 @@ class EasyUIBasedGenerateTaskPipeline(BasedGenerateTaskPipeline[EasyUIAppGenerat
 
     def _process_stream_response(
         self, publisher: AppGeneratorTTSPublisher | None, trace_manager: TraceQueueManager | None = None
-    ) -> Generator[StreamResponse, None, None]:
+    ) -> Generator[StreamResponse]:
         """
         Process stream response.
         :return:

api/core/base/tts/audio_mime.py

@@ -148,9 +148,9 @@ class AudioStream(Iterator[bytes]):
     """Own a MIME-checked stream, including resources opened while peeking."""
 
     def __init__(
-        self, chunks: Generator[bytes, None, None], *, source: Iterable[_AudioChunk], iterator: Iterator[_AudioChunk]
+        self, chunks: Generator[bytes], *, source: Iterable[_AudioChunk], iterator: Iterator[_AudioChunk]
     ) -> None:
-        self._chunks: Generator[bytes, None, None] = chunks
+        self._chunks: Generator[bytes] = chunks
         self._source: Iterable[_AudioChunk] = source
         self._iterator: Iterator[_AudioChunk] = iterator
         self._closed: bool = False
@@ -209,7 +209,7 @@ def inspect_audio_stream(
             close_stream(audio_stream)
         raise
 
-    def validated_stream() -> Generator[bytes, None, None]:
+    def validated_stream() -> Generator[bytes]:
         for chunk in chain(leading_chunks, iterator):
             audio, chunk_mime_type = _extract_audio_chunk(chunk)
             normalized_chunk_mime_type = _normalize_reported_mime_type(chunk_mime_type, "chunk")

api/core/callback_handler/workflow_tool_callback_handler.py

@@ -18,7 +18,7 @@ class DifyWorkflowCallbackHandler(DifyAgentCallbackHandler):
         message_id: str | None = None,
         timer: Any | None = None,
         trace_manager: TraceQueueManager | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         for tool_output in tool_outputs:
             if dify_config.DEBUG:
                 print_text("\n[on_tool_execution]\n", color=self.color)

api/core/datasource/datasource_manager.py

@@ -169,7 +169,7 @@ class DatasourceManager:
                 doc_runtime.runtime.credentials = dict(credentials)
             if datasource_param is None:
                 raise ValueError("datasource_param is required for ONLINE_DOCUMENT streaming")
-            inner_gen: Generator[DatasourceMessage, None, None] = doc_runtime.get_online_document_page_content(
+            inner_gen: Generator[DatasourceMessage] = doc_runtime.get_online_document_page_content(
                 user_id=user_id,
                 datasource_parameters=GetOnlineDocumentPageContentRequest(
                     workspace_id=datasource_param.workspace_id,
@@ -216,7 +216,7 @@ class DatasourceManager:
         variable_pool: Any,
         datasource_param: DatasourceParameter | None = None,
         online_drive_request: OnlineDriveDownloadFileParam | None = None,
-    ) -> Generator[StreamChunkEvent | StreamCompletedEvent, None, None]:
+    ) -> Generator[StreamChunkEvent | StreamCompletedEvent]:
         ds_type = DatasourceProviderType.value_of(datasource_type)
 
         messages = cls.stream_online_results(

api/core/datasource/online_document/online_document_plugin.py

@@ -36,7 +36,7 @@ class OnlineDocumentDatasourcePlugin(DatasourcePlugin):
         user_id: str,
         datasource_parameters: dict[str, Any],
         provider_type: str,
-    ) -> Generator[OnlineDocumentPagesMessage, None, None]:
+    ) -> Generator[OnlineDocumentPagesMessage]:
         manager = PluginDatasourceManager()
 
         return manager.get_online_document_pages(
@@ -54,7 +54,7 @@ class OnlineDocumentDatasourcePlugin(DatasourcePlugin):
         user_id: str,
         datasource_parameters: GetOnlineDocumentPageContentRequest,
         provider_type: str,
-    ) -> Generator[DatasourceMessage, None, None]:
+    ) -> Generator[DatasourceMessage]:
         manager = PluginDatasourceManager()
 
         return manager.get_online_document_page_content(

api/core/datasource/online_drive/online_drive_plugin.py

@@ -37,7 +37,7 @@ class OnlineDriveDatasourcePlugin(DatasourcePlugin):
         user_id: str,
         request: OnlineDriveBrowseFilesRequest,
         provider_type: str,
-    ) -> Generator[OnlineDriveBrowseFilesResponse, None, None]:
+    ) -> Generator[OnlineDriveBrowseFilesResponse]:
         manager = PluginDatasourceManager()
 
         return manager.online_drive_browse_files(
@@ -55,7 +55,7 @@ class OnlineDriveDatasourcePlugin(DatasourcePlugin):
         user_id: str,
         request: OnlineDriveDownloadFileRequest,
         provider_type: str,
-    ) -> Generator[DatasourceMessage, None, None]:
+    ) -> Generator[DatasourceMessage]:
         manager = PluginDatasourceManager()
 
         return manager.online_drive_download_file(

api/core/datasource/utils/message_transformer.py

@@ -15,11 +15,11 @@ class DatasourceFileMessageTransformer:
     @classmethod
     def transform_datasource_invoke_messages(
         cls,
-        messages: Generator[DatasourceMessage, None, None],
+        messages: Generator[DatasourceMessage],
         user_id: str,
         tenant_id: str,
         conversation_id: str | None = None,
-    ) -> Generator[DatasourceMessage, None, None]:
+    ) -> Generator[DatasourceMessage]:
         """
         Transform datasource message and handle file download
         """

api/core/datasource/website_crawl/website_crawl_plugin.py

@@ -34,7 +34,7 @@ class WebsiteCrawlDatasourcePlugin(DatasourcePlugin):
         user_id: str,
         datasource_parameters: Mapping[str, Any],
         provider_type: str,
-    ) -> Generator[WebsiteCrawlMessage, None, None]:
+    ) -> Generator[WebsiteCrawlMessage]:
         manager = PluginDatasourceManager()
 
         return manager.get_website_crawl(

api/core/llm_generator/output_parser/structured_output.py

@@ -57,7 +57,7 @@ def invoke_llm_with_structured_output(
     stream: Literal[True],
     callbacks: list[Callback] | None = None,
     request_metadata: Mapping[str, object] | None = None,
-) -> Generator[LLMResultChunkWithStructuredOutput, None, None]: ...
+) -> Generator[LLMResultChunkWithStructuredOutput]: ...
 @overload
 def invoke_llm_with_structured_output(
     *,
@@ -87,7 +87,7 @@ def invoke_llm_with_structured_output(
     stream: bool = True,
     callbacks: list[Callback] | None = None,
     request_metadata: Mapping[str, object] | None = None,
-) -> LLMResultWithStructuredOutput | Generator[LLMResultChunkWithStructuredOutput, None, None]: ...
+) -> LLMResultWithStructuredOutput | Generator[LLMResultChunkWithStructuredOutput]: ...
 def invoke_llm_with_structured_output(
     *,
     provider: str,
@@ -101,7 +101,7 @@ def invoke_llm_with_structured_output(
     stream: bool = True,
     callbacks: list[Callback] | None = None,
     request_metadata: Mapping[str, object] | None = None,
-) -> LLMResultWithStructuredOutput | Generator[LLMResultChunkWithStructuredOutput, None, None]:
+) -> LLMResultWithStructuredOutput | Generator[LLMResultChunkWithStructuredOutput]:
     """
     Invoke large language model with structured output
     1. This method invokes model_instance.invoke_llm with json_schema
@@ -162,7 +162,7 @@ def invoke_llm_with_structured_output(
         )
     else:
 
-        def generator() -> Generator[LLMResultChunkWithStructuredOutput, None, None]:
+        def generator() -> Generator[LLMResultChunkWithStructuredOutput]:
             result_text: str = ""
             prompt_messages: Sequence[PromptMessage] = []
             system_fingerprint: str | None = None

api/core/mcp/client/sse_client.py

@@ -261,7 +261,7 @@ def sse_client(
     headers: dict[str, Any] | None = None,
     timeout: float = 5.0,
     sse_read_timeout: float = 1 * 60,
-) -> Generator[tuple[ReadQueue, WriteQueue], None, None]:
+) -> Generator[tuple[ReadQueue, WriteQueue]]:
     """
     Client transport for SSE.
     `sse_read_timeout` determines how long (in seconds) the client will wait for a new
@@ -352,7 +352,7 @@ def send_message(http_client: httpx.Client, endpoint_url: str, session_message:
 
 def read_messages(
     sse_client: SSEClient,
-) -> Generator[SessionMessage | Exception, None, None]:
+) -> Generator[SessionMessage | Exception]:
     """
     Read messages from the SSE client.
 

api/core/mcp/client/streamable_client.py

@@ -491,8 +491,6 @@ def streamablehttp_client(
         ClientToServerQueue,  # Queue for sending messages TO server
         GetSessionIdCallback,
     ],
-    None,
-    None,
 ]:
     """
     Client transport for StreamableHTTP.

api/core/mcp/utils.py

@@ -127,9 +127,7 @@ def ssrf_proxy_sse_connect(url: str, **kwargs) -> AbstractContextManager[httpx_s
         raise
 
 
-def create_mcp_error_response(
-    request_id: int | str | None, code: int, message: str, data=None
-) -> Generator[bytes, None, None]:
+def create_mcp_error_response(request_id: int | str | None, code: int, message: str, data=None) -> Generator[bytes]:
     """Create MCP error response"""
     error_data = ErrorData(code=code, message=message, data=data)
     json_response = JSONRPCError(

api/core/model_context.py

@@ -21,7 +21,7 @@ def get_credit_usage_metadata() -> Mapping[str, object] | None:
 
 
 @contextmanager
-def use_credit_usage_metadata(metadata: Mapping[str, object] | None) -> Generator[None, None, None]:
+def use_credit_usage_metadata(metadata: Mapping[str, object] | None) -> Generator[None]:
     if metadata is None:
         yield
         return

api/core/model_manager.py

@@ -803,7 +803,7 @@ class QuotaManagedModelInstance(ModelInstance):
         content_text: str,
         voice: str,
         request_metadata: Mapping[str, object] | None = None,
-    ) -> Generator[bytes, None, None]:
+    ) -> Generator[bytes]:
         effective_request_metadata = self._resolve_request_metadata(request_metadata)
         reservation = self._reserve_quota_for_request(effective_request_metadata)
         try:

api/core/plugin/backwards_invocation/app.py

@@ -79,7 +79,7 @@ class PluginAppBackwardsInvocation(BaseBackwardsInvocation):
         files: list[dict],
         session: Session,
         end_users: AppScopedEndUserProvisioner,
-    ) -> Generator[Mapping | str, None, None] | Mapping:
+    ) -> Generator[Mapping | str] | Mapping:
         """
         invoke app
         """
@@ -122,7 +122,7 @@ class PluginAppBackwardsInvocation(BaseBackwardsInvocation):
         inputs: Mapping,
         files: list[dict],
         session: Session,
-    ) -> Generator[Mapping | str, None, None] | Mapping:
+    ) -> Generator[Mapping | str] | Mapping:
         """
         invoke chat app
         """
@@ -193,7 +193,7 @@ class PluginAppBackwardsInvocation(BaseBackwardsInvocation):
         stream: bool,
         inputs: Mapping,
         files: list[dict],
-    ) -> Generator[Mapping | str, None, None] | Mapping:
+    ) -> Generator[Mapping | str] | Mapping:
         """
         invoke workflow app
         """
@@ -222,7 +222,7 @@ class PluginAppBackwardsInvocation(BaseBackwardsInvocation):
         inputs: Mapping,
         files: list[dict],
         session: Session,
-    ) -> Generator[Mapping | str, None, None] | Mapping:
+    ) -> Generator[Mapping | str] | Mapping:
         """
         invoke completion app
         """

api/core/plugin/backwards_invocation/base.py

@@ -5,7 +5,7 @@ from pydantic import BaseModel
 
 class BaseBackwardsInvocation:
     @classmethod
-    def convert_to_event_stream(cls, response: Generator[BaseModel | Mapping | str, None, None] | BaseModel | Mapping):
+    def convert_to_event_stream(cls, response: Generator[BaseModel | Mapping | str] | BaseModel | Mapping):
         if isinstance(response, Generator):
             try:
                 for chunk in response:

api/core/plugin/backwards_invocation/model.py

@@ -58,7 +58,7 @@ class PluginModelBackwardsInvocation(BaseBackwardsInvocation):
     @classmethod
     def invoke_llm(
         cls, user_id: str, tenant: Tenant, payload: RequestInvokeLLM
-    ) -> Generator[LLMResultChunk, None, None] | LLMResult:
+    ) -> Generator[LLMResultChunk] | LLMResult:
         """
         invoke llm
         """
@@ -81,7 +81,7 @@ class PluginModelBackwardsInvocation(BaseBackwardsInvocation):
 
         if isinstance(response, Generator):
 
-            def handle() -> Generator[LLMResultChunk, None, None]:
+            def handle() -> Generator[LLMResultChunk]:
                 for chunk in response:
                     chunk.prompt_messages = []
                     yield chunk
@@ -89,7 +89,7 @@ class PluginModelBackwardsInvocation(BaseBackwardsInvocation):
             return handle()
         else:
 
-            def handle_non_streaming(response: LLMResult) -> Generator[LLMResultChunk, None, None]:
+            def handle_non_streaming(response: LLMResult) -> Generator[LLMResultChunk]:
                 yield LLMResultChunk(
                     model=response.model,
                     prompt_messages=[],
@@ -138,7 +138,7 @@ class PluginModelBackwardsInvocation(BaseBackwardsInvocation):
 
         if isinstance(response, Generator):
 
-            def handle() -> Generator[LLMResultChunkWithStructuredOutput, None, None]:
+            def handle() -> Generator[LLMResultChunkWithStructuredOutput]:
                 for chunk in response:
                     chunk.prompt_messages = []
                     yield chunk
@@ -148,7 +148,7 @@ class PluginModelBackwardsInvocation(BaseBackwardsInvocation):
 
             def handle_non_streaming(
                 response: LLMResultWithStructuredOutput,
-            ) -> Generator[LLMResultChunkWithStructuredOutput, None, None]:
+            ) -> Generator[LLMResultChunkWithStructuredOutput]:
                 yield LLMResultChunkWithStructuredOutput(
                     model=response.model,
                     prompt_messages=[],
@@ -218,7 +218,7 @@ class PluginModelBackwardsInvocation(BaseBackwardsInvocation):
             model=payload.model,
         )
 
-        def handle() -> Generator[dict[str, Any], None, None]:
+        def handle() -> Generator[dict[str, Any]]:
             response = model_instance.invoke_tts(content_text=payload.content_text, voice=payload.voice)
             audio_stream, mime_type = inspect_audio_stream(response, get_model_audio_mime_type(model_instance))
             for chunk in audio_stream:

api/core/plugin/backwards_invocation/tool.py

@@ -27,7 +27,7 @@ class PluginToolBackwardsInvocation(BaseBackwardsInvocation):
         tool_name: str,
         tool_parameters: dict[str, Any],
         credential_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         """
         invoke tool
         """

api/core/plugin/impl/agent.py

@@ -86,7 +86,7 @@ class PluginAgentClient(BasePluginClient):
         app_id: str | None = None,
         message_id: str | None = None,
         context: PluginInvokeContext | None = None,
-    ) -> Generator[AgentInvokeMessage, None, None]:
+    ) -> Generator[AgentInvokeMessage]:
         """
         Invoke the agent with the given tenant, user, plugin, provider, name and parameters.
         """

api/core/plugin/impl/base.py

@@ -79,7 +79,7 @@ _httpx_client: httpx.Client = get_pooled_http_client(
 
 
 @contextmanager
-def use_plugin_daemon_request_timeout(timeout_seconds: float) -> Generator[None, None, None]:
+def use_plugin_daemon_request_timeout(timeout_seconds: float) -> Generator[None]:
     """Temporarily shorten plugin-daemon requests made in the current context."""
     if timeout_seconds <= 0:
         raise ValueError("timeout_seconds must be greater than zero")
@@ -227,7 +227,7 @@ class BasePluginClient:
         headers: dict[str, str] | None = None,
         data: bytes | dict[str, Any] | None = None,
         files: dict[str, Any] | None = None,
-    ) -> Generator[str, None, None]:
+    ) -> Generator[str]:
         """
         Make a stream request to the plugin daemon inner API
         """
@@ -270,7 +270,7 @@ class BasePluginClient:
         data: bytes | dict[str, Any] | None = None,
         params: dict[str, Any] | None = None,
         files: dict[str, Any] | None = None,
-    ) -> Generator[T, None, None]:
+    ) -> Generator[T]:
         """
         Make a stream request to the plugin daemon inner API and yield the response as a model.
         """
@@ -358,7 +358,7 @@ class BasePluginClient:
         data: bytes | dict[str, Any] | None = None,
         params: dict[str, Any] | None = None,
         files: dict[str, Any] | None = None,
-    ) -> Generator[T, None, None]:
+    ) -> Generator[T]:
         """
         Make a stream request to the plugin daemon inner API and yield the response as a model.
         """

api/core/plugin/impl/datasource.py

@@ -144,7 +144,7 @@ class PluginDatasourceManager(BasePluginClient):
         credentials: dict[str, Any],
         datasource_parameters: Mapping[str, Any],
         provider_type: str,
-    ) -> Generator[WebsiteCrawlMessage, None, None]:
+    ) -> Generator[WebsiteCrawlMessage]:
         """
         Invoke the datasource with the given tenant, user, plugin, provider, name, credentials and parameters.
         """
@@ -179,7 +179,7 @@ class PluginDatasourceManager(BasePluginClient):
         credentials: dict[str, Any],
         datasource_parameters: Mapping[str, Any],
         provider_type: str,
-    ) -> Generator[OnlineDocumentPagesMessage, None, None]:
+    ) -> Generator[OnlineDocumentPagesMessage]:
         """
         Invoke the datasource with the given tenant, user, plugin, provider, name, credentials and parameters.
         """
@@ -214,7 +214,7 @@ class PluginDatasourceManager(BasePluginClient):
         credentials: dict[str, Any],
         datasource_parameters: GetOnlineDocumentPageContentRequest,
         provider_type: str,
-    ) -> Generator[DatasourceMessage, None, None]:
+    ) -> Generator[DatasourceMessage]:
         """
         Invoke the datasource with the given tenant, user, plugin, provider, name, credentials and parameters.
         """
@@ -249,7 +249,7 @@ class PluginDatasourceManager(BasePluginClient):
         credentials: dict[str, Any],
         request: OnlineDriveBrowseFilesRequest,
         provider_type: str,
-    ) -> Generator[OnlineDriveBrowseFilesResponse, None, None]:
+    ) -> Generator[OnlineDriveBrowseFilesResponse]:
         """
         Invoke the datasource with the given tenant, user, plugin, provider, name, credentials and parameters.
         """
@@ -285,7 +285,7 @@ class PluginDatasourceManager(BasePluginClient):
         credentials: dict[str, Any],
         request: OnlineDriveDownloadFileRequest,
         provider_type: str,
-    ) -> Generator[DatasourceMessage, None, None]:
+    ) -> Generator[DatasourceMessage]:
         """
         Invoke the datasource with the given tenant, user, plugin, provider, name, credentials and parameters.
         """

api/core/plugin/impl/model.py

@@ -181,7 +181,7 @@ class PluginModelClient(BasePluginClient):
         stop: list[str] | None = None,
         stream: bool = True,
         app_id: str | None = None,
-    ) -> Generator[LLMResultChunk, None, None]:
+    ) -> Generator[LLMResultChunk]:
         """
         Invoke llm
         """
@@ -582,7 +582,7 @@ class PluginModelClient(BasePluginClient):
         credentials: dict[str, Any],
         content_text: str,
         voice: str,
-    ) -> Generator[TTSAudioChunk, None, None]:
+    ) -> Generator[TTSAudioChunk]:
         """
         Invoke tts
         """

api/core/plugin/impl/model_runtime.py

@@ -77,7 +77,7 @@ class _PluginStructuredOutputModelInstance:
         stop: Sequence[str] | None = None,
         stream: bool = True,
         callbacks: object | None = None,
-    ) -> LLMResult | Generator[LLMResultChunk, None, None]:
+    ) -> LLMResult | Generator[LLMResultChunk]:
         del callbacks
         if stream:
             return self._runtime.invoke_llm(
@@ -303,7 +303,7 @@ class PluginModelRuntime(ModelRuntime):
         stop: Sequence[str] | None,
         stream: Literal[True],
         request_metadata: Mapping[str, object] | None = None,
-    ) -> Generator[LLMResultChunk, None, None]: ...
+    ) -> Generator[LLMResultChunk]: ...
 
     @override
     def invoke_llm(
@@ -318,7 +318,7 @@ class PluginModelRuntime(ModelRuntime):
         stop: Sequence[str] | None,
         stream: bool,
         request_metadata: Mapping[str, object] | None = None,
-    ) -> LLMResult | Generator[LLMResultChunk, None, None]:
+    ) -> LLMResult | Generator[LLMResultChunk]:
         app_id = request_metadata.get("app_id") if request_metadata else None
         if not isinstance(app_id, str):
             app_id = None
@@ -387,7 +387,7 @@ class PluginModelRuntime(ModelRuntime):
         prompt_messages: Sequence[PromptMessage],
         stop: Sequence[str] | None,
         stream: Literal[True],
-    ) -> Generator[LLMResultChunkWithStructuredOutput, None, None]: ...
+    ) -> Generator[LLMResultChunkWithStructuredOutput]: ...
 
     @override
     def invoke_llm_with_structured_output(
@@ -401,7 +401,7 @@ class PluginModelRuntime(ModelRuntime):
         prompt_messages: Sequence[PromptMessage],
         stop: Sequence[str] | None,
         stream: bool,
-    ) -> LLMResultWithStructuredOutput | Generator[LLMResultChunkWithStructuredOutput, None, None]:
+    ) -> LLMResultWithStructuredOutput | Generator[LLMResultChunkWithStructuredOutput]:
         model_schema = self.get_model_schema(
             provider=provider,
             model_type=ModelType.LLM,

api/core/plugin/impl/tool.py

@@ -94,7 +94,7 @@ class PluginToolManager(BasePluginClient):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         """
         Invoke the tool with the given tenant, user, plugin, provider, name, credentials and parameters.
         """

api/core/plugin/impl/trigger.py

@@ -95,7 +95,7 @@ class PluginTriggerClient(BasePluginClient):
         Invoke a trigger with the given parameters.
         """
         provider_id = TriggerProviderID(provider)
-        response: Generator[TriggerInvokeEventResponse, None, None] = self._request_with_plugin_daemon_response_stream(
+        response: Generator[TriggerInvokeEventResponse] = self._request_with_plugin_daemon_response_stream(
             method="POST",
             path=f"plugin/{tenant_id}/dispatch/trigger/invoke_event",
             type_=TriggerInvokeEventResponse,
@@ -130,7 +130,7 @@ class PluginTriggerClient(BasePluginClient):
         Validate the credentials of the trigger provider.
         """
         provider_id = TriggerProviderID(provider)
-        response: Generator[TriggerValidateProviderCredentialsResponse, None, None] = (
+        response: Generator[TriggerValidateProviderCredentialsResponse] = (
             self._request_with_plugin_daemon_response_stream(
                 method="POST",
                 path=f"plugin/{tenant_id}/dispatch/trigger/validate_credentials",
@@ -205,7 +205,7 @@ class PluginTriggerClient(BasePluginClient):
         Subscribe to a trigger.
         """
         provider_id = TriggerProviderID(provider)
-        response: Generator[TriggerSubscriptionResponse, None, None] = self._request_with_plugin_daemon_response_stream(
+        response: Generator[TriggerSubscriptionResponse] = self._request_with_plugin_daemon_response_stream(
             method="POST",
             path=f"plugin/{tenant_id}/dispatch/trigger/subscribe",
             type_=TriggerSubscriptionResponse,
@@ -243,7 +243,7 @@ class PluginTriggerClient(BasePluginClient):
         Unsubscribe from a trigger.
         """
         provider_id = TriggerProviderID(provider)
-        response: Generator[TriggerSubscriptionResponse, None, None] = self._request_with_plugin_daemon_response_stream(
+        response: Generator[TriggerSubscriptionResponse] = self._request_with_plugin_daemon_response_stream(
             method="POST",
             path=f"plugin/{tenant_id}/dispatch/trigger/unsubscribe",
             type_=TriggerSubscriptionResponse,
@@ -280,7 +280,7 @@ class PluginTriggerClient(BasePluginClient):
         Refresh a trigger subscription.
         """
         provider_id = TriggerProviderID(provider)
-        response: Generator[TriggerSubscriptionResponse, None, None] = self._request_with_plugin_daemon_response_stream(
+        response: Generator[TriggerSubscriptionResponse] = self._request_with_plugin_daemon_response_stream(
             method="POST",
             path=f"plugin/{tenant_id}/dispatch/trigger/refresh",
             type_=TriggerSubscriptionResponse,

api/core/plugin/utils/chunk_merger.py

@@ -20,10 +20,10 @@ class FileChunk:
 
 
 def merge_blob_chunks[T: ToolInvokeMessage | AgentInvokeMessage](
-    response: Generator[T, None, None],
+    response: Generator[T],
     max_file_size: int = 30 * 1024 * 1024,
     max_chunk_size: int = 8192,
-) -> Generator[T, None, None]:
+) -> Generator[T]:
     """
     Merge streaming blob chunks into complete blob messages.
 

api/core/rag/extractor/blob/blob.py

@@ -77,7 +77,7 @@ class Blob(BaseModel):
                 raise ValueError(f"Unable to get bytes for blob {self}")
 
     @contextlib.contextmanager
-    def as_bytes_io(self) -> Generator[BytesIO | BufferedReader, None, None]:
+    def as_bytes_io(self) -> Generator[BytesIO | BufferedReader]:
         """Read data as a byte stream."""
         match self.data:
             case bytes():

api/core/rag/extractor/watercrawl/provider.py

@@ -123,7 +123,7 @@ class WaterCrawlProvider:
 
     def _get_results(
         self, crawl_request_id: str, query_params: dict[str, Any] | None = None
-    ) -> Generator[WatercrawlDocumentData, None, None]:
+    ) -> Generator[WatercrawlDocumentData]:
         page = 0
         page_size = 100
 

api/core/rag/index_processor/index_processor_base.py

@@ -1,6 +1,5 @@
 """Abstract interface for document loader implementations."""
 
-import cgi
 import logging
 import mimetypes
 import os
@@ -13,6 +12,7 @@ from urllib.parse import unquote, urlparse
 import httpx
 from sqlalchemy import select
 from sqlalchemy.orm import Session
+from werkzeug.http import parse_options_header
 
 from configs import dify_config
 from core.entities.knowledge_entities import PreviewDetail
@@ -278,10 +278,10 @@ class BaseIndexProcessor(ABC):
 
             content_disposition = response.headers.get("content-disposition")
             if content_disposition:
-                _, params = cgi.parse_header(content_disposition)
+                _, params = parse_options_header(content_disposition)
                 if "filename" in params:
+                    # Werkzeug decodes filename*; preserve literal percent escapes in the result.
                     filename = params["filename"]
-                    filename = unquote(filename)
 
             if not filename:
                 parsed_url = urlparse(image_url)

api/core/rag/retrieval/dataset_retrieval.py

@@ -1643,7 +1643,7 @@ class DatasetRetrieval:
         try:
             # handle invoke result
             invoke_result = cast(
-                Generator[LLMResult, None, None],
+                Generator[LLMResult],
                 model_instance.invoke_llm(
                     prompt_messages=prompt_messages,
                     model_parameters=model_config.parameters,

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

@@ -160,7 +160,7 @@ class ReactMultiDatasetRouter:
             model_type=ModelType.LLM,
             model=model_instance.model_name,
         )
-        invoke_result: Generator[LLMResult, None, None] = bound_model_instance.invoke_llm(
+        invoke_result: Generator[LLMResult] = bound_model_instance.invoke_llm(
             prompt_messages=prompt_messages,
             model_parameters=completion_param,
             stop=stop,

api/core/tools/__base/tool.py

@@ -72,13 +72,13 @@ class Tool(ABC):
         match result:
             case ToolInvokeMessage():
 
-                def single_generator() -> Generator[ToolInvokeMessage, None, None]:
+                def single_generator() -> Generator[ToolInvokeMessage]:
                     yield result
 
                 return single_generator()
             case list():
 
-                def generator() -> Generator[ToolInvokeMessage, None, None]:
+                def generator() -> Generator[ToolInvokeMessage]:
                     yield from result
 
                 return generator()
@@ -106,7 +106,7 @@ class Tool(ABC):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> ToolInvokeMessage | list[ToolInvokeMessage] | Generator[ToolInvokeMessage, None, None]:
+    ) -> ToolInvokeMessage | list[ToolInvokeMessage] | Generator[ToolInvokeMessage]:
         pass
 
     def get_runtime_parameters(

api/core/tools/builtin_tool/providers/audio/tools/asr.py

@@ -28,7 +28,7 @@ class ASRTool(BuiltinTool):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         if not self.runtime:
             raise ValueError("Runtime is required")
         runtime = self.runtime

api/core/tools/builtin_tool/providers/audio/tools/tts.py

@@ -26,7 +26,7 @@ class TTSTool(BuiltinTool):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         if not self.runtime:
             raise ValueError("Runtime is required")
         runtime = self.runtime

api/core/tools/builtin_tool/providers/code/tools/simple_code.py

@@ -19,7 +19,7 @@ class SimpleCode(BuiltinTool):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         """
         invoke simple code
         """

api/core/tools/builtin_tool/providers/time/tools/current_time.py

@@ -19,7 +19,7 @@ class CurrentTimeTool(BuiltinTool):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         """
         invoke tools
         """

api/core/tools/builtin_tool/providers/time/tools/localtime_to_timestamp.py

@@ -20,7 +20,7 @@ class LocaltimeToTimestampTool(BuiltinTool):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         """
         Convert localtime to timestamp
         """

api/core/tools/builtin_tool/providers/time/tools/timestamp_to_localtime.py

@@ -20,7 +20,7 @@ class TimestampToLocaltimeTool(BuiltinTool):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         """
         Convert timestamp to localtime
         """

api/core/tools/builtin_tool/providers/time/tools/timezone_conversion.py

@@ -20,7 +20,7 @@ class TimezoneConversionTool(BuiltinTool):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         """
         Convert time to equivalent time zone
         """

api/core/tools/builtin_tool/providers/time/tools/weekday.py

@@ -19,7 +19,7 @@ class WeekdayTool(BuiltinTool):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         """
         Calculate the day of the week for a given date
         """

api/core/tools/builtin_tool/providers/webscraper/tools/webscraper.py

@@ -19,7 +19,7 @@ class WebscraperTool(BuiltinTool):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         """
         invoke tools
         """

api/core/tools/custom_tool/tool.py

@@ -387,7 +387,7 @@ class ApiTool(Tool):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         """
         invoke http request
         """

api/core/tools/mcp_tool/tool.py

@@ -74,7 +74,7 @@ class MCPTool(Tool):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         result = self.invoke_remote_mcp_tool(tool_parameters, user_id=user_id, app_id=app_id)
 
         # Extract usage metadata from MCP protocol's _meta field
@@ -111,7 +111,7 @@ class MCPTool(Tool):
         elif result.structuredContent:
             yield self.create_json_message(result.structuredContent)
 
-    def _process_text_content(self, content: TextContent) -> Generator[ToolInvokeMessage, None, None]:
+    def _process_text_content(self, content: TextContent) -> Generator[ToolInvokeMessage]:
         """Process text content and yield appropriate messages."""
         # Check if content looks like JSON before attempting to parse
         text = content.text.strip()
@@ -126,7 +126,7 @@ class MCPTool(Tool):
         # If not JSON or parsing failed, treat as plain text
         yield self.create_text_message(content.text)
 
-    def _process_json_content(self, content_json: Any) -> Generator[ToolInvokeMessage, None, None]:
+    def _process_json_content(self, content_json: Any) -> Generator[ToolInvokeMessage]:
         """Process JSON content based on its type."""
         match content_json:
             case dict():
@@ -137,7 +137,7 @@ class MCPTool(Tool):
                 # For primitive types (str, int, bool, etc.), convert to string
                 yield self.create_text_message(str(content_json))
 
-    def _process_json_list(self, json_list: list) -> Generator[ToolInvokeMessage, None, None]:
+    def _process_json_list(self, json_list: list) -> Generator[ToolInvokeMessage]:
         """Process a list of JSON items."""
         if any(not isinstance(item, dict) for item in json_list):
             # If the list contains any non-dict item, treat the entire list as a text message.

api/core/tools/plugin_tool/tool.py

@@ -35,7 +35,7 @@ class PluginTool(Tool):
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         manager = PluginToolManager()
 
         tool_parameters = convert_parameters_to_plugin_format(tool_parameters)

api/core/tools/tool_engine.py

@@ -88,7 +88,7 @@ class ToolEngine:
 
             def message_callback(
                 invocation_meta_dict: dict[str, ToolInvokeMeta],
-                messages: Generator[ToolInvokeMessage | ToolInvokeMeta, None, None],
+                messages: Generator[ToolInvokeMessage | ToolInvokeMeta],
             ):
                 for message in messages:
                     if isinstance(message, ToolInvokeMeta):
@@ -167,7 +167,7 @@ class ToolEngine:
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage, None, None]:
+    ) -> Generator[ToolInvokeMessage]:
         """
         Workflow invokes the tool with the given arguments.
         """
@@ -211,7 +211,7 @@ class ToolEngine:
         conversation_id: str | None = None,
         app_id: str | None = None,
         message_id: str | None = None,
-    ) -> Generator[ToolInvokeMessage | ToolInvokeMeta, None, None]:
+    ) -> Generator[ToolInvokeMessage | ToolInvokeMeta]:
         """
         Invoke the tool with the given arguments.
         """
@@ -288,7 +288,7 @@ class ToolEngine:
     @staticmethod
     def _extract_tool_response_binary_and_text(
         tool_response: list[ToolInvokeMessage],
-    ) -> Generator[ToolInvokeMessageBinary, None, None]:
+    ) -> Generator[ToolInvokeMessageBinary]:
         """
         Extract tool response binary
         """

api/core/tools/tool_manager.py

@@ -606,7 +606,7 @@ class ToolManager:
     @classmethod
     def list_builtin_providers(
         cls, tenant_id: str
-    ) -> Generator[BuiltinToolProviderController | PluginToolProviderController, None, None]:
+    ) -> Generator[BuiltinToolProviderController | PluginToolProviderController]:
         """
         list all the builtin providers
         """
@@ -615,7 +615,7 @@ class ToolManager:
         yield from cls.list_plugin_providers(tenant_id)
 
     @classmethod
-    def _list_hardcoded_providers(cls) -> Generator[BuiltinToolProviderController, None, None]:
+    def _list_hardcoded_providers(cls) -> Generator[BuiltinToolProviderController]:
         """
         list all the builtin providers
         """