[pre-commit.ci] pre-commit suggestions (#266)

* [pre-commit.ci] pre-commit suggestions

updates:
- [github.com/astral-sh/ruff-pre-commit: v0.5.6 → v0.6.3](astral-sh/ruff-pre-commit@v0.5.6...v0.6.3)

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Author: pre-commit-ci[bot]

.pre-commit-config.yaml

@@ -59,7 +59,7 @@ repos:
           - pydocstyle
 
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.5.6
+    rev: v0.6.3
     hooks:
       - id: ruff-format
         args: ["--preview"]

tests/conftest.py

@@ -83,35 +83,35 @@ def unbatch(self, output):
         yield from output
 
 
-@pytest.fixture()
+@pytest.fixture
 def simple_litapi():
     return SimpleLitAPI()
 
 
-@pytest.fixture()
+@pytest.fixture
 def simple_stream_api():
     return SimpleStreamAPI()
 
 
-@pytest.fixture()
+@pytest.fixture
 def simple_batched_stream_api():
     return SimpleBatchedStreamAPI()
 
 
-@pytest.fixture()
+@pytest.fixture
 def lit_server(simple_litapi):
     server = LitServer(simple_litapi, accelerator="cpu", devices=1, timeout=10)
     with wrap_litserve_start(server) as s:
         yield s
 
 
-@pytest.fixture()
+@pytest.fixture
 def sync_testclient(lit_server):
     with TestClient(lit_server.app) as client:
         yield client
 
 
-@pytest.fixture()
+@pytest.fixture
 def killall():
     def _run(process):
         parent = psutil.Process(process.pid)
@@ -122,7 +122,7 @@ def _run(process):
     return _run
 
 
-@pytest.fixture()
+@pytest.fixture
 def openai_request_data():
     return {
         "model": "",
@@ -139,7 +139,7 @@ def openai_request_data():
     }
 
 
-@pytest.fixture()
+@pytest.fixture
 def openai_response_data():
     return {
         "id": "chatcmpl-9dEtoQu4g45g3431SZ2s98S",
@@ -164,7 +164,7 @@ def openai_response_data():
     }
 
 
-@pytest.fixture()
+@pytest.fixture
 def openai_request_data_with_image():
     return {
         "model": "lit",
@@ -192,7 +192,7 @@ def openai_request_data_with_image():
     }
 
 
-@pytest.fixture()
+@pytest.fixture
 def openai_request_data_with_tools():
     return {
         "model": "lit",
@@ -226,7 +226,7 @@ def openai_request_data_with_tools():
     }
 
 
-@pytest.fixture()
+@pytest.fixture
 def openai_request_data_with_response_format():
     return {
         "model": "lit",

tests/test_batch.py

@@ -82,7 +82,7 @@ def encode_response(self, output) -> Response:
         return {"output": float(output)}
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_batched():
     api = SimpleBatchLitAPI()
     server = LitServer(api, accelerator="cpu", devices=1, timeout=10, max_batch_size=2, batch_timeout=4)
@@ -97,7 +97,7 @@ async def test_batched():
     assert response2.json() == {"output": 11.0}
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_unbatched():
     api = SimpleTorchAPI()
     server = LitServer(api, accelerator="cpu", devices=1, timeout=10, max_batch_size=1)

tests/test_examples.py

@@ -14,7 +14,7 @@
 import litserve as ls
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_simple_pytorch_api():
     api = ls.test_examples.SimpleTorchAPI()
     server = ls.LitServer(api, accelerator="cpu")
@@ -24,7 +24,7 @@ async def test_simple_pytorch_api():
             assert response.json() == {"output": 9.0}
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_simple_batched_api():
     api = ls.test_examples.SimpleBatchedAPI()
     server = ls.LitServer(api, max_batch_size=4, batch_timeout=0.1)
@@ -34,7 +34,7 @@ async def test_simple_batched_api():
             assert response.json() == {"output": 16.0}
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_simple_api():
     api = ls.test_examples.SimpleLitAPI()
     server = ls.LitServer(api)
@@ -44,15 +44,15 @@ async def test_simple_api():
             assert response.json() == {"output": 16.0}
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_simple_api_without_server():
     api = ls.test_examples.SimpleLitAPI()
     api.setup(None)
     assert api.model is not None, "Model should be loaded after setup"
     assert api.predict(4) == 16, "Model should be able to predict"
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_simple_pytorch_api_without_server():
     api = ls.test_examples.SimpleTorchAPI()
     api.setup("cpu")
@@ -63,7 +63,7 @@ async def test_simple_pytorch_api_without_server():
     assert api.encode_response(9) == {"output": 9}, "Response should be encoded"
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_simple_stream_api_without_server():
     api = SimpleStreamAPI()
     api.setup(None)
@@ -77,7 +77,7 @@ async def test_simple_stream_api_without_server():
     ], "Response should be encoded"
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_openai_with_usage():
     api = OpenAIWithUsage()
     api.setup(None)
@@ -93,7 +93,7 @@ async def test_openai_with_usage():
     ], "Response should match expected output"
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_openai_with_usage_encode_response():
     api = OpenAIWithUsageEncodeResponse()
     api.setup(None)
@@ -114,7 +114,7 @@ async def test_openai_with_usage_encode_response():
     ], "Encoded response should match expected output"
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_openai_batching_with_usage():
     api = OpenAIBatchingWithUsage()
     api.setup(None)
@@ -136,7 +136,7 @@ async def test_openai_batching_with_usage():
     ], "Encoded batched response should match expected output"
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_openai_batch_context():
     api = OpenAIBatchContext()
     api.setup(None)

tests/test_lit_server.py

@@ -60,7 +60,7 @@ def test_device_identifiers(lifespan_mock, simple_litapi):
     assert server.devices[1][0] == "cuda:2"
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_stream(simple_stream_api):
     server = LitServer(simple_stream_api, stream=True, timeout=10)
     expected_output1 = "prompt=Hello generated_output=LitServe is streaming output".lower().replace(" ", "")
@@ -81,7 +81,7 @@ async def test_stream(simple_stream_api):
             ), "Server returns input prompt and generated output which didn't match."
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_batched_stream_server(simple_batched_stream_api):
     server = LitServer(simple_batched_stream_api, stream=True, max_batch_size=4, batch_timeout=2, timeout=30)
     expected_output1 = "Hello LitServe is streaming output".lower().replace(" ", "")
@@ -315,7 +315,7 @@ def encode_response(self, output, context):
         return {"output": input}
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 @patch("litserve.server.load_and_raise")
 async def test_inject_context(mocked_load_and_raise):
     def dummy_load_and_raise(resp):

tests/test_loops.py

@@ -28,7 +28,7 @@
 from litserve.utils import LitAPIStatus
 
 
-@pytest.fixture()
+@pytest.fixture
 def loop_args():
     requests_queue = Queue()
     requests_queue.put((0, "uuid-123", time.monotonic(), 1))  # response_queue_id, uid, timestamp, x_enc

tests/test_simple.py

@@ -119,7 +119,7 @@ def unbatch(self, output):
         return list(output)
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_timeout():
     # Scenario: first request completes, second request times out in queue
     api = SlowLitAPI()  # takes 2 seconds for each prediction
@@ -146,7 +146,7 @@ async def test_timeout():
             ), "Server takes longer than specified timeout and request should timeout"
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_batch_timeout():
     # Scenario: first 2 requests finish as a batch and third request times out in queue
     server = LitServer(

tests/test_specs.py

@@ -30,7 +30,7 @@
 from litserve.specs.openai import OpenAISpec, ChatMessage
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_openai_spec(openai_request_data):
     spec = OpenAISpec()
     server = ls.LitServer(TestAPI(), spec=spec)
@@ -45,7 +45,7 @@ async def test_openai_spec(openai_request_data):
 
 
 # OpenAIWithUsage
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 @pytest.mark.parametrize(
     ("api", "batch_size"),
     [
@@ -72,7 +72,7 @@ async def test_openai_token_usage(api, batch_size, openai_request_data, openai_r
             assert result["usage"] == openai_response_data["usage"]
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_openai_spec_with_image(openai_request_data_with_image):
     server = ls.LitServer(TestAPI(), spec=OpenAISpec())
     with wrap_litserve_start(server) as server:
@@ -85,7 +85,7 @@ async def test_openai_spec_with_image(openai_request_data_with_image):
             ), "LitAPI predict response should match with the generated output"
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_override_encode(openai_request_data):
     server = ls.LitServer(TestAPIWithCustomEncode(), spec=OpenAISpec())
     with wrap_litserve_start(server) as server:
@@ -98,7 +98,7 @@ async def test_override_encode(openai_request_data):
             ), "LitAPI predict response should match with the generated output"
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_openai_spec_with_tools(openai_request_data_with_tools):
     spec = OpenAISpec()
     server = ls.LitServer(TestAPIWithToolCalls(), spec=spec)
@@ -118,7 +118,7 @@ async def test_openai_spec_with_tools(openai_request_data_with_tools):
             ], "LitAPI predict response should match with the generated output"
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_openai_spec_with_response_format(openai_request_data_with_response_format):
     spec = OpenAISpec()
     server = ls.LitServer(TestAPIWithStructuredOutput(), spec=spec)
@@ -148,7 +148,7 @@ def encode_response(self, output):
         return ChatMessage(role="assistant", content="This is a generated output")
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_openai_spec_validation(openai_request_data):
     server = ls.LitServer(IncorrectAPI1(), spec=OpenAISpec())
     with pytest.raises(ValueError, match="predict is not a generator"), wrap_litserve_start(server) as server:
@@ -172,7 +172,7 @@ def predict(self, prompt, context):
                 return
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_oai_prepopulated_context(openai_request_data):
     openai_request_data["max_tokens"] = 3
     spec = OpenAISpec()
@@ -194,7 +194,7 @@ def predict(self, prompt):
         raise HTTPException(501, "test LitAPI.predict error")
 
 
-@pytest.mark.asyncio()
+@pytest.mark.asyncio
 async def test_fail_http(openai_request_data):
     server = ls.LitServer(WrongLitAPI(), spec=ls.OpenAISpec())
     with wrap_litserve_start(server) as server: