Ruff 🐶

tests/conftest.py

@@ -1,6 +1,6 @@
 import asyncio
 import contextlib
-from asyncio import AbstractEventLoop, get_running_loop
+from asyncio import AbstractEventLoop
 from functools import partial
 from importlib.metadata import version
 from random import getrandbits
@@ -199,11 +199,6 @@ async def app(
         yield manager.app
 
 
-@pytest.fixture
-async def loop() -> AbstractEventLoop:
-    return get_running_loop()
-
-
 @pytest.fixture
 def httpx_clients(
     app: ASGIApp,

tests/datasets/test_experiments.py

@@ -5,7 +5,6 @@
 from unittest.mock import patch
 
 import httpx
-import pytest
 from sqlalchemy import select
 from strawberry.relay import GlobalID
 
@@ -37,8 +36,8 @@ async def test_run_experiment(
     simple_dataset: Any,
     dialect: str,
 ) -> None:
-    if dialect == "postgresql":
-        pytest.xfail("TODO: Convert this to an integration test")
+    # if dialect == "postgresql":
+    #     pytest.xfail("TODO: Convert this to an integration test")
 
     async with db() as session:
         nonexistent_experiment = (await session.execute(select(models.Experiment))).scalar()
@@ -103,16 +102,18 @@ def experiment_task(_) -> Dict[str, str]:
 
         # Wait until all evaluations are complete
         async def wait_for_evaluations():
-            timeout = 10
+            timeout = 30
             interval = 0.5
             total_wait = 0
             while total_wait < timeout:
                 async with db() as session:
                     evaluations = (
                         (
                             await session.execute(
-                                select(models.ExperimentRunAnnotation)
-                                .where(models.ExperimentRunAnnotation.experiment_run_id == experiment_id)
+                                select(models.ExperimentRunAnnotation).where(
+                                    models.ExperimentRunAnnotation.experiment_run_id
+                                    == experiment_id
+                                )
                             )
                         )
                         .scalars()
@@ -124,6 +125,7 @@ async def wait_for_evaluations():
                 total_wait += interval
             else:
                 raise TimeoutError("Evaluations did not complete in time")
+
         await wait_for_evaluations()
 
         experiment_model = (await session.execute(select(models.Experiment))).scalar()
@@ -178,8 +180,8 @@ async def test_run_experiment_with_llm_eval(
     simple_dataset: Any,
     dialect: str,
 ) -> None:
-    if dialect == "postgresql":
-        pytest.xfail("This test fails on PostgreSQL")
+    # if dialect == "postgresql":
+    #     pytest.xfail("This test fails on PostgreSQL")
 
     async with db() as session:
         nonexistent_experiment = (await session.execute(select(models.Experiment))).scalar()
@@ -290,8 +292,8 @@ async def test_run_evaluation(
     simple_dataset_with_one_experiment_run: Any,
     dialect: str,
 ) -> None:
-    if dialect == "postgresql":
-        pytest.xfail("This test fails on PostgreSQL")
+    # if dialect == "postgresql":
+    #     pytest.xfail("This test fails on PostgreSQL")
 
     experiment = Experiment(
         id=str(GlobalID("Experiment", "0")),

tests/server/api/routers/v1/test_spans.py

@@ -20,9 +20,6 @@ async def test_span_round_tripping_with_docs(
     dialect: str,
     span_data_with_documents: Any,
 ) -> None:
-    # if dialect == "sqlite":
-    #     pytest.xfail("TODO: Convert this to an integration test")
-
     df = cast(pd.DataFrame, px_client.get_spans_dataframe())
     new_ids = {span_id: getrandbits(64).to_bytes(8, "big").hex() for span_id in df.index}
     for span_id_col_name in ("context.span_id", "parent_id"):