Remove warnings from optional models

workflowai/core/client/agent.py

@@ -472,6 +472,7 @@ async def reply(
         user_message: Optional[str] = None,
         tool_results: Optional[Iterable[ToolCallResult]] = None,
         current_iteration: int = 0,
+        max_retries: int = 2,
         **kwargs: Unpack[RunParams[AgentOutput]],
     ):
         """Reply to a run to provide additional information or context.
@@ -489,7 +490,18 @@ async def reply(
         prepared_run = await self._prepare_reply(run_id, user_message, tool_results, stream=False, **kwargs)
         validator, new_kwargs = self._sanitize_validator(kwargs, self._default_validator)
 
-        res = await self.api.post(prepared_run.route, prepared_run.request, returns=RunResponse, run=True)
+        async def _with_retries():
+            err: Optional[WorkflowAIError] = None
+            for _ in range(max_retries):
+                try:
+                    return await self.api.post(prepared_run.route, prepared_run.request, returns=RunResponse, run=True)
+                except WorkflowAIError as e:  # noqa: PERF203
+                    if e.code != "object_not_found":
+                        raise e
+                    err = e
+            raise err or RuntimeError("This should never raise")
+
+        res = await _with_retries()
         return await self._build_run(
             res,
             prepared_run.schema_id,

workflowai/core/client/agent_test.py

@@ -1078,3 +1078,75 @@ async def test_stream_validation_final_error(
 
         assert e.value.partial_output == {"message": 1}
         assert e.value.run_id == "1"
+
+
+class TestReply:
+    async def test_reply_success(self, httpx_mock: HTTPXMock, agent: Agent[HelloTaskInput, HelloTaskOutput]):
+        httpx_mock.add_response(
+            url="http://localhost:8000/v1/_/agents/123/runs/1/reply",
+            json=fixtures_json("task_run.json"),
+        )
+        reply = await agent.reply(run_id="1", user_message="test message")
+        assert reply.output.message == "Austin"
+
+        assert len(httpx_mock.get_requests()) == 1
+
+    async def test_reply_first_404(self, httpx_mock: HTTPXMock, agent: Agent[HelloTaskInput, HelloTaskOutput]):
+        """Check that we retry once if the run is not found"""
+
+        httpx_mock.add_response(
+            url="http://localhost:8000/v1/_/agents/123/runs/1/reply",
+            status_code=404,
+            json={
+                "error": {
+                    "code": "object_not_found",
+                },
+            },
+        )
+
+        httpx_mock.add_response(
+            url="http://localhost:8000/v1/_/agents/123/runs/1/reply",
+            json=fixtures_json("task_run.json"),
+        )
+
+        reply = await agent.reply(run_id="1", user_message="test message")
+        assert reply.output.message == "Austin"
+
+        assert len(httpx_mock.get_requests()) == 2
+
+    async def test_reply_not_not_found_error(
+        self,
+        httpx_mock: HTTPXMock,
+        agent: Agent[HelloTaskInput, HelloTaskOutput],
+    ):
+        """Check that we raise the error if it's not a 404"""
+        httpx_mock.add_response(
+            url="http://localhost:8000/v1/_/agents/123/runs/1/reply",
+            status_code=400,
+            json={
+                "error": {
+                    "code": "whatever",
+                },
+            },
+        )
+        with pytest.raises(WorkflowAIError) as e:
+            await agent.reply(run_id="1", user_message="test message")
+        assert e.value.code == "whatever"
+        assert len(httpx_mock.get_requests()) == 1
+
+    async def test_reply_multiple_retries(self, httpx_mock: HTTPXMock, agent: Agent[HelloTaskInput, HelloTaskOutput]):
+        """Check that we retry once if the run is not found"""
+        httpx_mock.add_response(
+            url="http://localhost:8000/v1/_/agents/123/runs/1/reply",
+            status_code=404,
+            json={
+                "error": {
+                    "code": "object_not_found",
+                },
+            },
+            is_reusable=True,
+        )
+        with pytest.raises(WorkflowAIError) as e:
+            await agent.reply(run_id="1", user_message="test message")
+        assert e.value.code == "object_not_found"
+        assert len(httpx_mock.get_requests()) == 2

workflowai/core/utils/_pydantic.py

@@ -1,3 +1,4 @@
+import warnings
 from collections.abc import Mapping, Sequence
 from typing import Any, TypeVar, get_args, get_origin
 
@@ -25,9 +26,11 @@ def _copy_field_info(field_info: FieldInfo, **overrides: Any):
     certain values.
     """
 
+    _excluded = {"annotation", "required"}
+
     kwargs = overrides
     for k, v in field_info.__repr_args__():
-        if k in kwargs or not k:
+        if k in kwargs or not k or k in _excluded:
             continue
         kwargs[k] = v
 
@@ -79,7 +82,6 @@ def partial_model(base: type[BM]) -> type[BM]:
         overrides: dict[str, Any] = {}
         try:
             annotation = _optional_annotation(field.annotation)
-            overrides["annotation"] = annotation
             overrides["default"] = _default_value_from_annotation(annotation)
         except Exception:  # noqa: BLE001
             logger.debug("Failed to make annotation optional", exc_info=True)
@@ -95,10 +97,12 @@ def custom_eq(o1: BM, o2: Any):
             return False
         return o1.model_dump() == o2.model_dump()
 
-    return create_model(  # pyright: ignore [reportCallIssue, reportUnknownVariableType]
-        f"Partial{base.__name__}",
-        __base__=base,
-        __eq__=custom_eq,
-        __hash__=base.__hash__,
-        **default_fields,  # pyright: ignore [reportArgumentType]
-    )
+    with warnings.catch_warnings():
+        warnings.filterwarnings("ignore", category=RuntimeWarning, message="fields may not start with an underscore")
+        return create_model(  # pyright: ignore [reportCallIssue, reportUnknownVariableType]
+            f"Partial{base.__name__}",
+            __base__=base,
+            __eq__=custom_eq,
+            __hash__=base.__hash__,
+            **default_fields,  # pyright: ignore [reportArgumentType]
+        )

workflowai/core/utils/_pydantic_test.py

@@ -7,7 +7,7 @@
 
 
 class TestPartialModel:
-    def test_partial_model_equals(self):
+    def test_partial_model_equals(self, recwarn: pytest.WarningsRecorder):
         class SimpleModel(BaseModel):
             name: str
 
@@ -16,6 +16,8 @@ class SimpleModel(BaseModel):
 
         assert SimpleModel(name="John") == partial.model_validate({"name": "John"})
 
+        assert len(recwarn.list) == 0
+
     def test_simple_model(self):
         class SimpleModel(BaseModel):
             name1: str