Migrate linter from pylint to ruff (#1665)

tests/async_engine/api_server_async_engine.py

@@ -14,7 +14,6 @@ app = vllm.entrypoints.api_server.app
 
 class AsyncLLMEngineWithStats(AsyncLLMEngine):
 
-    # pylint: disable=redefined-outer-name
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
         self._num_aborts = 0

tests/async_engine/test_api_server.py

@@ -24,7 +24,6 @@ def _query_server(prompt: str) -> dict:
 def api_server():
     script_path = Path(__file__).parent.joinpath(
         "api_server_async_engine.py").absolute()
-    # pylint: disable=consider-using-with
     uvicorn_process = subprocess.Popen([
         sys.executable, "-u",
         str(script_path), "--model", "facebook/opt-125m"
@@ -33,7 +32,6 @@ def api_server():
     uvicorn_process.terminate()
 
 
-# pylint: disable=redefined-outer-name, unused-argument
 def test_api_server(api_server):
     """
     Run the API server and test it.
@@ -49,11 +47,10 @@ def test_api_server(api_server):
         prompts = ["Hello world"] * 1
         result = None
         while not result:
-            # pylint: disable=bare-except
             try:
-                for result in pool.map(_query_server, prompts):
+                for _ in pool.map(_query_server, prompts):
                     break
-            except:
+            except Exception:
                 time.sleep(1)
 
         # Actual tests start here

tests/conftest.py

@@ -8,7 +8,6 @@ from vllm import LLM, SamplingParams
 from vllm.transformers_utils.tokenizer import get_tokenizer
 
 _TEST_PROMPTS = [
-    # pylint: disable=line-too-long
     "vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs.",
     "Briefly describe the major milestones in the development of artificial intelligence from 1950 to 2020.",
     "Compare and contrast artificial intelligence with human intelligence in terms of processing information.",

tests/engine/test_detokenize.py

@@ -5,10 +5,9 @@ from transformers import AutoTokenizer
 from vllm.transformers_utils.tokenizer import detokenize_incrementally
 
 TRUTH = [
-    # pylint: disable=line-too-long
-    "Hello here, this is a simple test",
-    "vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. It is designed to be used in production environments, where inference and serving",
-    "我很感谢你的热情"
+    "Hello here, this is a simple test",  # noqa: E501
+    "vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. It is designed to be used in production environments, where inference and serving",  # noqa: E501
+    "我很感谢你的热情"  # noqa: E501
 ]
 TOKENIZERS = [
     "facebook/opt-125m",

tests/kernels/test_attention.py

@@ -211,7 +211,7 @@ def test_paged_attention(
             alibi_slopes,
         )
     else:
-        assert False, f"Unknown version: {version}"
+        raise AssertionError(f"Unknown version: {version}")
 
     # Run the reference implementation.
     ref_output = torch.empty_like(query)

tests/samplers/test_sampler.py

@@ -1,4 +1,3 @@
-# pylint: disable=protected-access
 import random
 from typing import Tuple
 from unittest.mock import patch
@@ -20,10 +19,10 @@ class MockLogitsSampler(Sampler):
 
     def forward(self, *args, **kwargs):
         with patch("vllm.model_executor.layers.sampler._prune_hidden_states",
-                   lambda x, y: x):
-            with patch("vllm.model_executor.layers.sampler._get_logits",
+                   lambda x, y: x), patch(
+                       "vllm.model_executor.layers.sampler._get_logits",
                        lambda *args, **kwargs: self.fake_logits):
-                return super().forward(*args, **kwargs)
+            return super().forward(*args, **kwargs)
 
 
 def _prepare_test(
@@ -214,6 +213,6 @@ def test_sampler_logits_processors(seed: int):
     sampler_output = sampler(embedding=None,
                              hidden_states=input_tensor,
                              input_metadata=input_metadata)
-    for i, sequence_output in enumerate(sampler_output):
+    for _, sequence_output in enumerate(sampler_output):
         for idx, nth_output in enumerate(sequence_output.samples):
             assert nth_output.output_token == idx

tests/worker/test_worker.py

@@ -1,4 +1,3 @@
-# pylint: disable=protected-access
 import random
 import torch