Merge remote-tracking branch 'origin/main' into feat/evaluation

api/configs/middleware/__init__.py

@@ -160,6 +160,16 @@ class DatabaseConfig(BaseSettings):
         default="",
     )
 
+    DB_SESSION_TIMEZONE_OVERRIDE: str = Field(
+        description=(
+            "PostgreSQL session timezone override injected via startup options."
+            " Default is 'UTC' for out-of-the-box consistency."
+            " Set to empty string to disable app-level timezone injection, for example when using RDS Proxy"
+            " together with a database-side default timezone."
+        ),
+        default="UTC",
+    )
+
     @computed_field  # type: ignore[prop-decorator]
     @property
     def SQLALCHEMY_DATABASE_URI_SCHEME(self) -> str:
@@ -227,12 +237,13 @@ class DatabaseConfig(BaseSettings):
         connect_args: dict[str, str] = {}
         # Use the dynamic SQLALCHEMY_DATABASE_URI_SCHEME property
         if self.SQLALCHEMY_DATABASE_URI_SCHEME.startswith("postgresql"):
-            timezone_opt = "-c timezone=UTC"
-            if options:
-                merged_options = f"{options} {timezone_opt}"
-            else:
-                merged_options = timezone_opt
-            connect_args = {"options": merged_options}
+            merged_options = options.strip()
+            session_timezone_override = self.DB_SESSION_TIMEZONE_OVERRIDE.strip()
+            if session_timezone_override:
+                timezone_opt = f"-c timezone={session_timezone_override}"
+                merged_options = f"{merged_options} {timezone_opt}".strip() if merged_options else timezone_opt
+            if merged_options:
+                connect_args = {"options": merged_options}
 
         result: SQLAlchemyEngineOptionsDict = {
             "pool_size": self.SQLALCHEMY_POOL_SIZE,

api/core/agent/plugin_entities.py

@@ -84,7 +84,7 @@ class AgentStrategyEntity(BaseModel):
     identity: AgentStrategyIdentity
     parameters: list[AgentStrategyParameter] = Field(default_factory=list)
     description: I18nObject = Field(..., description="The description of the agent strategy")
-    output_schema: dict | None = None
+    output_schema: dict[str, Any] | None = None
     features: list[AgentFeature] | None = None
     meta_version: str | None = None
     # pydantic configs

api/core/app/app_config/common/sensitive_word_avoidance/manager.py

@@ -22,8 +22,8 @@ class SensitiveWordAvoidanceConfigManager:
 
     @classmethod
     def validate_and_set_defaults(
-        cls, tenant_id: str, config: dict, only_structure_validate: bool = False
-    ) -> tuple[dict, list[str]]:
+        cls, tenant_id: str, config: dict[str, Any], only_structure_validate: bool = False
+    ) -> tuple[dict[str, Any], list[str]]:
         if not config.get("sensitive_word_avoidance"):
             config["sensitive_word_avoidance"] = {"enabled": False}
 

api/core/app/app_config/easy_ui_based_app/model_config/manager.py

@@ -41,7 +41,7 @@ class ModelConfigManager:
         )
 
     @classmethod
-    def validate_and_set_defaults(cls, tenant_id: str, config: Mapping[str, Any]) -> tuple[dict, list[str]]:
+    def validate_and_set_defaults(cls, tenant_id: str, config: Mapping[str, Any]) -> tuple[dict[str, Any], list[str]]:
         """
         Validate and set defaults for model config
 

api/core/extension/api_based_extension_requestor.py

@@ -1,4 +1,4 @@
-from typing import cast
+from typing import Any, cast
 
 import httpx
 
@@ -14,7 +14,7 @@ class APIBasedExtensionRequestor:
         self.api_endpoint = api_endpoint
         self.api_key = api_key
 
-    def request(self, point: APIBasedExtensionPoint, params: dict):
+    def request(self, point: APIBasedExtensionPoint, params: dict[str, Any]) -> dict[str, Any]:
         """
         Request the api.
 
@@ -49,4 +49,4 @@ class APIBasedExtensionRequestor:
         if response.status_code != 200:
             raise ValueError(f"request error, status_code: {response.status_code}, content: {response.text[:100]}")
 
-        return cast(dict, response.json())
+        return cast(dict[str, Any], response.json())

api/core/extension/extensible.py

@@ -21,8 +21,8 @@ class ExtensionModule(StrEnum):
 class ModuleExtension(BaseModel):
     extension_class: Any | None = None
     name: str
-    label: dict | None = None
-    form_schema: list | None = None
+    label: dict[str, Any] | None = None
+    form_schema: list[dict[str, Any]] | None = None
     builtin: bool = True
     position: int | None = None
 

api/core/external_data_tool/factory.py

@@ -13,7 +13,7 @@ class ExternalDataToolFactory:
         )
 
     @classmethod
-    def validate_config(cls, name: str, tenant_id: str, config: dict[str, Any]):
+    def validate_config(cls, name: str, tenant_id: str, config: dict[str, Any]) -> None:
         """
         Validate the incoming form config data.
 

api/core/indexing_runner.py

@@ -737,7 +737,7 @@ class IndexingRunner:
     def _update_document_index_status(
         document_id: str,
         after_indexing_status: IndexingStatus,
-        extra_update_params: dict[Any, Any] | None = None,
+        extra_update_params: Mapping[Any, Any] | None = None,
     ):
         """
         Update the document indexing status.
@@ -764,7 +764,7 @@ class IndexingRunner:
         db.session.commit()
 
     @staticmethod
-    def _update_segments_by_document(dataset_document_id: str, update_params: dict[Any, Any]):
+    def _update_segments_by_document(dataset_document_id: str, update_params: Mapping[Any, Any]):
         """
         Update the document segment by document id.
         """

api/core/llm_generator/llm_generator.py

@@ -2,7 +2,7 @@ import json
 import logging
 import re
 from collections.abc import Sequence
-from typing import Protocol, TypedDict, cast
+from typing import Any, Protocol, TypedDict, cast
 
 import json_repair
 from graphon.enums import WorkflowNodeExecutionMetadataKey
@@ -533,7 +533,7 @@ class LLMGenerator:
     def __instruction_modify_common(
         tenant_id: str,
         model_config: ModelConfig,
-        last_run: dict | None,
+        last_run: dict[str, Any] | None,
         current: str | None,
         error_message: str | None,
         instruction: str,

api/core/llm_generator/output_parser/structured_output.py

@@ -202,7 +202,7 @@ def _handle_native_json_schema(
     structured_output_schema: Mapping,
     model_parameters: dict[str, Any],
     rules: list[ParameterRule],
-):
+) -> dict[str, Any]:
     """
     Handle structured output for models with native JSON schema support.
 
@@ -224,7 +224,7 @@ def _handle_native_json_schema(
     return model_parameters
 
 
-def _set_response_format(model_parameters: dict[str, Any], rules: list[ParameterRule]):
+def _set_response_format(model_parameters: dict[str, Any], rules: list[ParameterRule]) -> None:
     """
     Set the appropriate response format parameter based on model rules.
 
@@ -326,7 +326,7 @@ def _prepare_schema_for_model(provider: str, model_schema: AIModelEntity, schema
         return {"schema": processed_schema, "name": "llm_response"}
 
 
-def remove_additional_properties(schema: dict[str, Any]):
+def remove_additional_properties(schema: dict[str, Any]) -> None:
     """
     Remove additionalProperties fields from JSON schema.
     Used for models like Gemini that don't support this property.
@@ -349,7 +349,7 @@ def remove_additional_properties(schema: dict[str, Any]):
                     remove_additional_properties(item)
 
 
-def convert_boolean_to_string(schema: dict):
+def convert_boolean_to_string(schema: dict[str, Any]) -> None:
     """
     Convert boolean type specifications to string in JSON schema.
 

api/core/prompt/simple_prompt_transform.py

@@ -313,7 +313,7 @@ class SimplePromptTransform(PromptTransform):
 
         return prompt_message
 
-    def _get_prompt_rule(self, app_mode: AppMode, provider: str, model: str):
+    def _get_prompt_rule(self, app_mode: AppMode, provider: str, model: str) -> dict[str, Any]:
         """
         Get simple prompt rule.
         :param app_mode: app mode
@@ -325,7 +325,7 @@ class SimplePromptTransform(PromptTransform):
 
         # Check if the prompt file is already loaded
         if prompt_file_name in prompt_file_contents:
-            return cast(dict, prompt_file_contents[prompt_file_name])
+            return cast(dict[str, Any], prompt_file_contents[prompt_file_name])
 
         # Get the absolute path of the subdirectory
         prompt_path = os.path.join(os.path.dirname(os.path.realpath(__file__)), "prompt_templates")
@@ -338,7 +338,7 @@ class SimplePromptTransform(PromptTransform):
             # Store the content of the prompt file
             prompt_file_contents[prompt_file_name] = content
 
-            return cast(dict, content)
+            return cast(dict[str, Any], content)
 
     def _prompt_file_name(self, app_mode: AppMode, provider: str, model: str) -> str:
         # baichuan

api/core/rag/embedding/cached_embedding.py

@@ -106,7 +106,7 @@ class CacheEmbedding(Embeddings):
 
         return text_embeddings
 
-    def embed_multimodal_documents(self, multimodel_documents: list[dict]) -> list[list[float]]:
+    def embed_multimodal_documents(self, multimodel_documents: list[dict[str, Any]]) -> list[list[float]]:
         """Embed file documents."""
         # use doc embedding cache or store if not exists
         multimodel_embeddings: list[Any] = [None for _ in range(len(multimodel_documents))]

api/core/rag/embedding/embedding_base.py

@@ -11,7 +11,7 @@ class Embeddings(ABC):
         raise NotImplementedError
 
     @abstractmethod
-    def embed_multimodal_documents(self, multimodel_documents: list[dict]) -> list[list[float]]:
+    def embed_multimodal_documents(self, multimodel_documents: list[dict[str, Any]]) -> list[list[float]]:
         """Embed file documents."""
         raise NotImplementedError
 

api/core/rag/retrieval/output_parser/react_output.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 from dataclasses import dataclass
-from typing import NamedTuple, Union
+from typing import Any, NamedTuple, Union
 
 
 @dataclass
@@ -10,7 +10,7 @@ class ReactAction:
 
     tool: str
     """The name of the Tool to execute."""
-    tool_input: Union[str, dict]
+    tool_input: Union[str, dict[str, Any]]
     """The input to pass in to the Tool."""
     log: str
     """Additional information to log about the action."""
@@ -19,7 +19,7 @@ class ReactAction:
 class ReactFinish(NamedTuple):
     """The final return value of an ReactFinish."""
 
-    return_values: dict
+    return_values: dict[str, Any]
     """Dictionary of return values."""
     log: str
     """Additional information to log about the return value"""

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

@@ -1,5 +1,5 @@
 from collections.abc import Generator, Sequence
-from typing import Union
+from typing import Any, Union
 
 from graphon.model_runtime.entities.llm_entities import LLMResult, LLMUsage
 from graphon.model_runtime.entities.message_entities import PromptMessage, PromptMessageRole, PromptMessageTool
@@ -139,7 +139,7 @@ class ReactMultiDatasetRouter:
 
     def _invoke_llm(
         self,
-        completion_param: dict,
+        completion_param: dict[str, Any],
         model_instance: ModelInstance,
         prompt_messages: list[PromptMessage],
         stop: list[str],

api/core/rag/splitter/text_splitter.py

@@ -63,7 +63,7 @@ class TextSplitter(BaseDocumentTransformer, ABC):
     def split_text(self, text: str) -> list[str]:
         """Split text into multiple components."""
 
-    def create_documents(self, texts: list[str], metadatas: list[dict] | None = None) -> list[Document]:
+    def create_documents(self, texts: list[str], metadatas: list[dict[str, Any]] | None = None) -> list[Document]:
         """Create documents from a list of texts."""
         _metadatas = metadatas or [{}] * len(texts)
         documents = []

api/core/tools/entities/tool_bundle.py

@@ -1,4 +1,5 @@
 from collections.abc import Mapping
+from typing import Any
 
 from pydantic import BaseModel, Field
 
@@ -26,6 +27,6 @@ class ApiToolBundle(BaseModel):
     # icon
     icon: str | None = None
     # openapi operation
-    openapi: dict
+    openapi: dict[str, Any]
     # output schema
     output_schema: Mapping[str, object] = Field(default_factory=dict)

api/core/tools/tool_engine.py

@@ -47,7 +47,7 @@ class ToolEngine:
     @staticmethod
     def agent_invoke(
         tool: Tool,
-        tool_parameters: Union[str, dict],
+        tool_parameters: Union[str, dict[str, Any]],
         user_id: str,
         tenant_id: str,
         message: Message,
@@ -85,7 +85,7 @@ class ToolEngine:
             invocation_meta_dict: dict[str, ToolInvokeMeta] = {}
 
             def message_callback(
-                invocation_meta_dict: dict[str, Any],
+                invocation_meta_dict: dict[str, ToolInvokeMeta],
                 messages: Generator[ToolInvokeMessage | ToolInvokeMeta, None, None],
             ):
                 for message in messages:

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

@@ -1,4 +1,4 @@
-from typing import cast
+from typing import Any, cast
 
 from pydantic import BaseModel, Field
 from sqlalchemy import select
@@ -39,7 +39,7 @@ class DatasetRetrieverTool(DatasetRetrieverBaseTool):
     dataset_id: str
     user_id: str | None = None
     retrieve_config: DatasetRetrieveConfigEntity
-    inputs: dict
+    inputs: dict[str, Any]
 
     @classmethod
     def from_dataset(cls, dataset: Dataset, **kwargs):

api/core/tools/workflow_as_tool/tool.py

@@ -277,7 +277,7 @@ class WorkflowTool(Tool):
             session.expunge(app)
             return app
 
-    def _transform_args(self, tool_parameters: dict[str, Any]) -> tuple[dict[str, Any], list[dict[str, Any]]]:
+    def _transform_args(self, tool_parameters: dict[str, Any]) -> tuple[dict[str, Any], list[dict[str, str | None]]]:
         """
         transform the tool parameters
 
@@ -355,7 +355,7 @@ class WorkflowTool(Tool):
 
         return result, files
 
-    def _update_file_mapping(self, file_dict: dict[str, Any]):
+    def _update_file_mapping(self, file_dict: dict[str, Any]) -> dict[str, Any]:
         file_id = resolve_file_record_id(file_dict.get("reference") or file_dict.get("related_id"))
         transfer_method = FileTransferMethod.value_of(file_dict.get("transfer_method"))
         match transfer_method:

api/pyproject.toml

@@ -136,6 +136,9 @@ dify-vdb-weaviate = { workspace = true }
 [tool.uv]
 default-groups = ["storage", "tools", "vdb-all"]
 package = false
+override-dependencies = [
+    "pyarrow>=18.0.0",
+]
 
 [dependency-groups]
 

api/tests/unit_tests/configs/test_dify_config.py

@@ -145,7 +145,7 @@ def test_inner_api_config_exist(monkeypatch: pytest.MonkeyPatch):
 
 
 def test_db_extras_options_merging(monkeypatch: pytest.MonkeyPatch):
-    """Test that DB_EXTRAS options are properly merged with default timezone setting"""
+    """Test that DB_EXTRAS options are merged with the default timezone startup option."""
     # Set environment variables
     monkeypatch.setenv("DB_TYPE", "postgresql")
     monkeypatch.setenv("DB_USERNAME", "postgres")
@@ -158,15 +158,28 @@ def test_db_extras_options_merging(monkeypatch: pytest.MonkeyPatch):
     # Create config
     config = DifyConfig()
 
-    # Get engine options
-    engine_options = config.SQLALCHEMY_ENGINE_OPTIONS
-
-    # Verify options contains both search_path and timezone
-    options = engine_options["connect_args"]["options"]
+    options = config.SQLALCHEMY_ENGINE_OPTIONS["connect_args"]["options"]
     assert "search_path=myschema" in options
     assert "timezone=UTC" in options
 
 
+def test_db_session_timezone_override_can_disable_app_level_timezone_injection(monkeypatch: pytest.MonkeyPatch):
+    monkeypatch.setenv("DB_TYPE", "postgresql")
+    monkeypatch.setenv("DB_USERNAME", "postgres")
+    monkeypatch.setenv("DB_PASSWORD", "postgres")
+    monkeypatch.setenv("DB_HOST", "localhost")
+    monkeypatch.setenv("DB_PORT", "5432")
+    monkeypatch.setenv("DB_DATABASE", "dify")
+    monkeypatch.setenv("DB_EXTRAS", "options=-c search_path=myschema")
+    monkeypatch.setenv("DB_SESSION_TIMEZONE_OVERRIDE", "")
+
+    config = DifyConfig()
+
+    assert config.SQLALCHEMY_ENGINE_OPTIONS["connect_args"] == {
+        "options": "-c search_path=myschema",
+    }
+
+
 def test_pubsub_redis_url_default(monkeypatch: pytest.MonkeyPatch):
     os.environ.clear()
 

api/uv.lock

@@ -42,6 +42,7 @@ members = [
     "dify-vdb-vikingdb",
     "dify-vdb-weaviate",
 ]
+overrides = [{ name = "pyarrow", specifier = ">=18.0.0" }]
 
 [[package]]
 name = "abnf"
@@ -5356,20 +5357,17 @@ wheels = [
 
 [[package]]
 name = "pyarrow"
-version = "14.0.2"
+version = "23.0.1"
 source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "numpy" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/d7/8b/d18b7eb6fb22e5ed6ffcbc073c85dae635778dbd1270a6cf5d750b031e84/pyarrow-14.0.2.tar.gz", hash = "sha256:36cef6ba12b499d864d1def3e990f97949e0b79400d08b7cf74504ffbd3eb025", size = 1063645, upload-time = "2023-12-18T15:43:41.625Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/88/22/134986a4cc224d593c1afde5494d18ff629393d74cc2eddb176669f234a4/pyarrow-23.0.1.tar.gz", hash = "sha256:b8c5873e33440b2bc2f4a79d2b47017a89c5a24116c055625e6f2ee50523f019", size = 1167336, upload-time = "2026-02-16T10:14:12.39Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/69/5b/d8ab6c20c43b598228710e4e4a6cba03a01f6faa3d08afff9ce76fd0fd47/pyarrow-14.0.2-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:c87824a5ac52be210d32906c715f4ed7053d0180c1060ae3ff9b7e560f53f944", size = 26819585, upload-time = "2023-12-18T15:41:27.59Z" },
-    { url = "https://files.pythonhosted.org/packages/2d/29/bed2643d0dd5e9570405244a61f6db66c7f4704a6e9ce313f84fa5a3675a/pyarrow-14.0.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:a25eb2421a58e861f6ca91f43339d215476f4fe159eca603c55950c14f378cc5", size = 23965222, upload-time = "2023-12-18T15:41:32.449Z" },
-    { url = "https://files.pythonhosted.org/packages/2a/34/da464632e59a8cdd083370d69e6c14eae30221acb284f671c6bc9273fadd/pyarrow-14.0.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5c1da70d668af5620b8ba0a23f229030a4cd6c5f24a616a146f30d2386fec422", size = 35942036, upload-time = "2023-12-18T15:41:38.767Z" },
-    { url = "https://files.pythonhosted.org/packages/a8/ff/cbed4836d543b29f00d2355af67575c934999ff1d43e3f438ab0b1b394f1/pyarrow-14.0.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2cc61593c8e66194c7cdfae594503e91b926a228fba40b5cf25cc593563bcd07", size = 38089266, upload-time = "2023-12-18T15:41:47.617Z" },
-    { url = "https://files.pythonhosted.org/packages/38/41/345011cb831d3dbb2dab762fc244c745a5df94b199223a99af52a5f7dff6/pyarrow-14.0.2-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:78ea56f62fb7c0ae8ecb9afdd7893e3a7dbeb0b04106f5c08dbb23f9c0157591", size = 35404468, upload-time = "2023-12-18T15:41:54.49Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/af/2fc23ca2068ff02068d8dabf0fb85b6185df40ec825973470e613dbd8790/pyarrow-14.0.2-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:37c233ddbce0c67a76c0985612fef27c0c92aef9413cf5aa56952f359fcb7379", size = 38003134, upload-time = "2023-12-18T15:42:01.593Z" },
-    { url = "https://files.pythonhosted.org/packages/95/1f/9d912f66a87e3864f694e000977a6a70a644ea560289eac1d733983f215d/pyarrow-14.0.2-cp312-cp312-win_amd64.whl", hash = "sha256:e4b123ad0f6add92de898214d404e488167b87b5dd86e9a434126bc2b7a5578d", size = 25043754, upload-time = "2023-12-18T15:42:07.108Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/4b/4166bb5abbfe6f750fc60ad337c43ecf61340fa52ab386da6e8dbf9e63c4/pyarrow-23.0.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:f4b0dbfa124c0bb161f8b5ebb40f1a680b70279aa0c9901d44a2b5a20806039f", size = 34214575, upload-time = "2026-02-16T10:09:56.225Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/da/3f941e3734ac8088ea588b53e860baeddac8323ea40ce22e3d0baa865cc9/pyarrow-23.0.1-cp312-cp312-macosx_12_0_x86_64.whl", hash = "sha256:7707d2b6673f7de054e2e83d59f9e805939038eebe1763fe811ee8fa5c0cd1a7", size = 35832540, upload-time = "2026-02-16T10:10:03.428Z" },
+    { url = "https://files.pythonhosted.org/packages/88/7c/3d841c366620e906d54430817531b877ba646310296df42ef697308c2705/pyarrow-23.0.1-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:86ff03fb9f1a320266e0de855dee4b17da6794c595d207f89bba40d16b5c78b9", size = 44470940, upload-time = "2026-02-16T10:10:10.704Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/a5/da83046273d990f256cb79796a190bbf7ec999269705ddc609403f8c6b06/pyarrow-23.0.1-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:813d99f31275919c383aab17f0f455a04f5a429c261cc411b1e9a8f5e4aaaa05", size = 47586063, upload-time = "2026-02-16T10:10:17.95Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/3c/b7d2ebcff47a514f47f9da1e74b7949138c58cfeb108cdd4ee62f43f0cf3/pyarrow-23.0.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:bf5842f960cddd2ef757d486041d57c96483efc295a8c4a0e20e704cbbf39c67", size = 48173045, upload-time = "2026-02-16T10:10:25.363Z" },
+    { url = "https://files.pythonhosted.org/packages/43/b2/b40961262213beaba6acfc88698eb773dfce32ecdf34d19291db94c2bd73/pyarrow-23.0.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:564baf97c858ecc03ec01a41062e8f4698abc3e6e2acd79c01c2e97880a19730", size = 50621741, upload-time = "2026-02-16T10:10:33.477Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/70/1fdda42d65b28b078e93d75d371b2185a61da89dda4def8ba6ba41ebdeb4/pyarrow-23.0.1-cp312-cp312-win_amd64.whl", hash = "sha256:07deae7783782ac7250989a7b2ecde9b3c343a643f82e8a4df03d93b633006f0", size = 27620678, upload-time = "2026-02-16T10:10:39.31Z" },
 ]
 
 [[package]]