remove eager mode on feedback (#216)

Eager mode is not a preferred option for langsmith APIs. We thus change
the `/feedback` to use the regular submission of feedback flow.

langserve/server.py

@@ -34,7 +34,7 @@
 from langchain.schema.runnable import Runnable, RunnableConfig
 from langchain.schema.runnable.config import get_config_list, merge_configs
 from langsmith import client as ls_client
-from langsmith.utils import LangSmithNotFoundError, tracing_is_enabled
+from langsmith.utils import tracing_is_enabled
 from typing_extensions import Annotated
 
 try:
@@ -1046,6 +1046,11 @@ async def playground(
     async def feedback(feedback_create_req: FeedbackCreateRequest) -> Feedback:
         """
         Send feedback on an individual run to langsmith
+
+        Note that a successful response means that feedback was successfully
+        submitted. It does not guarantee that the feedback is recorded by
+        langsmith. Requests may be silently rejected if they are
+        unauthenticated or invalid by the server.
         """
 
         if not tracing_is_enabled() or not enable_feedback_endpoint:
@@ -1055,28 +1060,16 @@ async def feedback(feedback_create_req: FeedbackCreateRequest) -> Feedback:
                 + "enabled on your LangServe server.",
             )
 
-        try:
-            feedback_from_langsmith = langsmith_client.create_feedback(
-                feedback_create_req.run_id,
-                feedback_create_req.key,
-                score=feedback_create_req.score,
-                value=feedback_create_req.value,
-                comment=feedback_create_req.comment,
-                source_info={
-                    "from_langserve": True,
-                },
-                # We execute eagerly, meaning we confirm the run exists in
-                # LangSmith before returning a response to the user. This ensures
-                # that clients of the UI know that the feedback was successfully
-                # recorded before they receive a 200 response
-                eager=True,
-                # We lower the number of attempts to 3 to ensure we have time
-                # to wait for a run to show up, but do not take forever in cases
-                # of bad input
-                stop_after_attempt=3,
-            )
-        except LangSmithNotFoundError:
-            raise HTTPException(404, "No run with the given run_id exists")
+        feedback_from_langsmith = langsmith_client.create_feedback(
+            feedback_create_req.run_id,
+            feedback_create_req.key,
+            score=feedback_create_req.score,
+            value=feedback_create_req.value,
+            comment=feedback_create_req.comment,
+            source_info={
+                "from_langserve": True,
+            },
+        )
 
         # We purposefully select out fields from langsmith so that we don't
         # fail validation if langsmith adds extra fields. We prefer this over

tests/unit_tests/test_server_client.py

@@ -23,7 +23,6 @@
 from langchain.schema.runnable.base import RunnableLambda
 from langchain.schema.runnable.utils import ConfigurableField, Input, Output
 from langsmith import schemas as ls_schemas
-from langsmith.utils import LangSmithNotFoundError
 from pytest import MonkeyPatch
 from pytest_mock import MockerFixture
 from typing_extensions import TypedDict
@@ -1636,39 +1635,6 @@ async def test_feedback_succeeds_when_langsmith_enabled() -> None:
                 assert json_response == expected_response_json
 
 
-@pytest.mark.asyncio
-async def test_feedback_fails_when_run_doesnt_exist() -> None:
-    """Tests that the feedback endpoint can't accept feedback for a non-existent run."""
-
-    with patch("langserve.server.ls_client") as mocked_ls_client_package:
-        with patch("langserve.server.tracing_is_enabled") as tracing_is_enabled:
-            tracing_is_enabled.return_value = True
-            mocked_client = MagicMock(return_value=None)
-            mocked_ls_client_package.Client.return_value = mocked_client
-            mocked_client.create_feedback.side_effect = LangSmithNotFoundError(
-                "no run :/"
-            )
-            local_app = FastAPI()
-            add_routes(
-                local_app,
-                RunnableLambda(lambda foo: "hello"),
-                enable_feedback_endpoint=True,
-            )
-
-            async with get_async_test_client(
-                local_app, raise_app_exceptions=True
-            ) as async_client:
-                response = await async_client.post(
-                    "/feedback",
-                    json={
-                        "run_id": "f47ac10b-58cc-4372-a567-0e02b2c3d479",
-                        "key": "silliness",
-                        "score": 1000,
-                    },
-                )
-                assert response.status_code == 404
-
-
 @pytest.mark.asyncio
 async def test_feedback_fails_when_langsmith_disabled(app: FastAPI) -> None:
     """Tests that feedback is not sent to langsmith if langsmith is disabled."""