Add flake8-implicit-str-concat rules to Ruff (#33191)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2026-01-28 04:56:10 +00:00
committed by GitHub
parent a97b5e206d
commit 2eb673a088
16 changed files with 34 additions and 33 deletions
+1 -2
View File
@@ -134,8 +134,7 @@ class BenchmarkDataset(ABC):
content.append(mm_content)
else:
raise TypeError(
"Could not process multimodal content of type: "
+ f"{type(mm_content)}"
f"Could not process multimodal content of type: {type(mm_content)}"
)
return [{"role": "user", "content": content}]
+2 -2
View File
@@ -208,8 +208,8 @@ class TorchCompileWithNoGuardsWrapper:
if not hasattr(self._compiled_callable, "aot_compile"):
raise RuntimeError(
"aot_compile is not supported by the current configuration. "
+ "Please make sure torch.compile is enabled with the latest "
+ f"version of PyTorch (current using torch: {torch.__version__})"
"Please make sure torch.compile is enabled with the latest "
f"version of PyTorch (current using torch: {torch.__version__})"
)
return self._compiled_callable.aot_compile((args, kwargs))
@@ -406,8 +406,8 @@ class OpenAISpeechToText(OpenAIServing):
if request.response_format not in ["text", "json", "verbose_json"]:
return self.create_error_response(
("Currently only support response_format")
+ ("`text`, `json` or `verbose_json`")
"Currently only support response_format: "
"`text`, `json` or `verbose_json`"
)
if (
@@ -32,8 +32,8 @@ class AiterInt8ScaledMMLinearKernel(CutlassInt8ScaledMMLinearKernel):
return (
False,
"requires setting `VLLM_ROCM_USE_AITER=1` "
+ "and `VLLM_ROCM_USE_AITER_LINEAR=1`. "
+ "`VLLM_ROCM_USE_AITER_LINEAR` default is True.",
"and `VLLM_ROCM_USE_AITER_LINEAR=1`. "
"`VLLM_ROCM_USE_AITER_LINEAR` default is True.",
)
return True, None
@@ -97,9 +97,9 @@ class AiterInt8ScaledMMLinearKernel(CutlassInt8ScaledMMLinearKernel):
per_token_scale_a and per_channel_scale_b
), (
"Currently only support per-tensor-per-tensor GEMM "
+ " and per-token-per-channel GEMM through AITER"
" and per-token-per-channel GEMM through AITER"
" w8a8 scaled gemm. `AiterInt8ScaledMMLinearKernel` "
+ "does not support AITER block scaled GEMM."
"does not support AITER block scaled GEMM."
)
# gemm_a8w8_CK(a, b, scale_a, scale_b, bias) expects
+1 -1
View File
@@ -234,7 +234,7 @@ class Olmo3ReasoningParser(ReasoningParser):
# reasoning template.
reasoning_expr = (
rf"^(?:{self.think_start})?(?P<reasoning>.*?)"
+ rf"{self.think_end}(?P<content>.*)$"
rf"{self.think_end}(?P<content>.*)$"
)
self.reasoning_regex = re.compile(reasoning_expr, re.DOTALL)
@@ -216,7 +216,7 @@ class FlashAttnMLAMetadataBuilder(MLACommonMetadataBuilder[FlashAttnMLAMetadata]
# Ensure the persistent buffer is large enough
assert n <= self.scheduler_metadata.shape[0], (
f"Scheduler metadata size {n} exceeds buffer size "
+ f"{self.scheduler_metadata.shape[0]}"
f"{self.scheduler_metadata.shape[0]}"
)
self.scheduler_metadata[:n] = scheduler_metadata
# NOTE(woosuk): We should zero out the rest of the scheduler
+1 -1
View File
@@ -646,7 +646,7 @@ class ChunkedLocalAttentionManager(SingleTypeKVCacheManager):
"""
assert isinstance(kv_cache_spec, ChunkedLocalAttentionSpec), (
"ChunkedLocalAttentionManager can only be used for "
+ "chunked local attention groups"
"chunked local attention groups"
)
assert use_eagle is False, (
"Hybrid KV cache is not supported for " + "eagle + chunked local attention."
+1 -1
View File
@@ -618,7 +618,7 @@ class AsyncLLM(EngineClient):
except Exception as e2:
s = (
f"{e.__class__.__name__}: "
+ "error during printing an exception of class"
"error during printing an exception of class"
+ e2.__class__.__name__
)
logger.info("Request %s failed due to %s.", request_id, s)