[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
co-authored by
Claude Opus 5
parent
1085b64425
commit
651a88c09a
@@ -3,8 +3,7 @@
|
||||
"""The CLI entrypoints of vLLM.
|
||||
|
||||
Note that all future modules must be lazily loaded within main
|
||||
to avoid certain eager import breakage.
|
||||
"""
|
||||
to avoid certain eager import breakage."""
|
||||
|
||||
import importlib.metadata
|
||||
import sys
|
||||
|
||||
@@ -295,8 +295,7 @@ class GenerateBaseServing(BaseServing, BeamSearchOnlineMixin):
|
||||
) -> _T:
|
||||
"""Wrap a `create_*` coroutine so that, if it raises or returns an
|
||||
ErrorResponse (i.e. the request never reached the engine), the KV
|
||||
connector is notified to free any pinned remote-prefill blocks.
|
||||
"""
|
||||
connector is notified to free any pinned remote-prefill blocks."""
|
||||
kv_transfer_params = self.has_kv_connector and request.kv_transfer_params
|
||||
if not kv_transfer_params or not kv_transfer_params.get("do_remote_prefill"):
|
||||
return await awaitable
|
||||
|
||||
@@ -123,8 +123,7 @@ class OpenAIServingModels:
|
||||
|
||||
async def init_static_loras(self):
|
||||
"""Loads all static LoRA modules.
|
||||
Raises if any fail to load
|
||||
"""
|
||||
Raises if any fail to load"""
|
||||
if self.static_lora_modules is None:
|
||||
return
|
||||
for lora in self.static_lora_modules:
|
||||
@@ -149,8 +148,7 @@ class OpenAIServingModels:
|
||||
|
||||
async def show_available_models(self) -> ModelList:
|
||||
"""Show available models. This includes the base model and all
|
||||
adapters.
|
||||
"""
|
||||
adapters."""
|
||||
model_list = await self.registry.show_available_models()
|
||||
lora_cards = [
|
||||
ModelCard(
|
||||
|
||||
@@ -408,8 +408,7 @@ class ParsableContext(ConversationContext):
|
||||
|
||||
def need_builtin_tool_call(self) -> bool:
|
||||
"""Return true if the last message is a builtin tool call
|
||||
that the request has enabled.
|
||||
"""
|
||||
that the request has enabled."""
|
||||
last_message = self.response_messages[-1]
|
||||
if last_message.type != "function_call":
|
||||
return False
|
||||
|
||||
@@ -52,8 +52,7 @@ logger = init_logger(__name__)
|
||||
|
||||
def _parse_harmony_format_message(chat_msg: dict) -> Message:
|
||||
"""Reconstruct a Message from Harmony-format dict,
|
||||
preserving channel, recipient, and content_type.
|
||||
"""
|
||||
preserving channel, recipient, and content_type."""
|
||||
author_dict = chat_msg["author"]
|
||||
role = author_dict.get("role")
|
||||
name = author_dict.get("name")
|
||||
|
||||
@@ -117,8 +117,7 @@ def serialize_messages(msgs):
|
||||
|
||||
class ResponseRawMessageAndToken(OpenAIBaseModel):
|
||||
"""Class to show the raw message.
|
||||
If message / tokens diverge, tokens is the source of truth
|
||||
"""
|
||||
If message / tokens diverge, tokens is the source of truth"""
|
||||
|
||||
message: str
|
||||
tokens: list[int]
|
||||
|
||||
@@ -44,8 +44,7 @@ class PluginWithoutIOProcessorPlugins(PoolingIOProcessor):
|
||||
|
||||
class PluginWithIOProcessorPlugins(PoolingIOProcessor):
|
||||
"""IO Processor plugins are a feature that allows pre- and post-processing
|
||||
of the model input and output for pooling models.
|
||||
"""
|
||||
of the model input and output for pooling models."""
|
||||
|
||||
name = "plugin"
|
||||
|
||||
|
||||
@@ -1,8 +1,7 @@
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
|
||||
"""Encode/decode utilities for multimodal tensors and field metadata
|
||||
over JSON/HTTP, used by the disaggregated generate endpoint.
|
||||
"""
|
||||
over JSON/HTTP, used by the disaggregated generate endpoint."""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
|
||||
@@ -126,8 +126,7 @@ def _format_error(err: object) -> str:
|
||||
|
||||
def _is_internal_loc_segment(segment: str) -> bool:
|
||||
"""True if `segment` is a Pydantic-internal wrapper/union-branch
|
||||
marker rather than a user-meaningful field name or list index.
|
||||
"""
|
||||
marker rather than a user-meaningful field name or list index."""
|
||||
if _BRACKETED_INTERNAL_RE.search(segment):
|
||||
return True
|
||||
return segment.lower() in _INTERNAL_LOC_MARKERS
|
||||
|
||||
@@ -63,8 +63,7 @@ def _attach_router(app: FastAPI):
|
||||
@router.post("/v1/load_lora_adapter", dependencies=[Depends(validate_json_request)])
|
||||
async def load_lora_adapter(request: LoadLoRAAdapterRequest, raw_request: Request):
|
||||
"""Handle POST /v1/load_lora_adapter: load a LoRA adapter into the
|
||||
serving engine.
|
||||
"""
|
||||
serving engine."""
|
||||
handler: OpenAIServingModels = models(raw_request)
|
||||
response = await handler.load_lora_adapter(request)
|
||||
if isinstance(response, ErrorResponse):
|
||||
@@ -86,8 +85,7 @@ def _attach_router(app: FastAPI):
|
||||
request: UnloadLoRAAdapterRequest, raw_request: Request
|
||||
):
|
||||
"""Handle POST /v1/unload_lora_adapter: unload a LoRA adapter from
|
||||
the serving engine.
|
||||
"""
|
||||
the serving engine."""
|
||||
handler: OpenAIServingModels = models(raw_request)
|
||||
response = await handler.unload_lora_adapter(request)
|
||||
if isinstance(response, ErrorResponse):
|
||||
|
||||
@@ -30,8 +30,7 @@ EndpointFn = Callable[[RequestType, Request], Awaitable[Any]]
|
||||
def _snapshot_handler_levels() -> list[tuple[logging.Handler, int]]:
|
||||
"""Snapshot handler levels for loggers that may be affected by third-party
|
||||
logging configuration side effects (e.g. model_hosting_container_standards
|
||||
calling configure_root_logger() at import time).
|
||||
"""
|
||||
calling configure_root_logger() at import time)."""
|
||||
loggers = [logging.getLogger(), logging.getLogger("vllm")]
|
||||
seen: set[int] = set()
|
||||
snapshot: list[tuple[logging.Handler, int]] = []
|
||||
@@ -74,8 +73,7 @@ def _attach_router(
|
||||
model_config: ModelConfig | None = None,
|
||||
):
|
||||
"""Register the SageMaker hosting routes (/ping, /invocations) on the
|
||||
app.
|
||||
"""
|
||||
app."""
|
||||
import model_hosting_container_standards.sagemaker as sagemaker_standards
|
||||
|
||||
router = APIRouter()
|
||||
|
||||
@@ -33,8 +33,7 @@ def set_default_fingerprint_mode(
|
||||
custom_value: str | None = None,
|
||||
) -> None:
|
||||
"""Configure the fingerprint mode for subsequent ``get_system_fingerprint``
|
||||
calls. Called once at server startup.
|
||||
"""
|
||||
calls. Called once at server startup."""
|
||||
global _DEFAULT_MODE, _CUSTOM_VALUE
|
||||
_DEFAULT_MODE = mode
|
||||
_CUSTOM_VALUE = custom_value
|
||||
@@ -42,8 +41,7 @@ def set_default_fingerprint_mode(
|
||||
|
||||
def get_system_fingerprint(vllm_config: Any) -> str | None:
|
||||
"""Return the fingerprint for ``vllm_config`` using the mode configured by
|
||||
``set_default_fingerprint_mode``.
|
||||
"""
|
||||
``set_default_fingerprint_mode``."""
|
||||
return build_system_fingerprint(vllm_config, _DEFAULT_MODE, _CUSTOM_VALUE)
|
||||
|
||||
|
||||
|
||||
@@ -92,8 +92,7 @@ def asr_inter_chunk_separator(
|
||||
|
||||
class SpeechToTextBaseServing(GenerateBaseServing):
|
||||
"""Base class for speech-to-text operations like transcription and
|
||||
translation.
|
||||
"""
|
||||
translation."""
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
@@ -427,8 +426,7 @@ class SpeechToTextBaseServing(GenerateBaseServing):
|
||||
stream_generator_method: Callable[..., AsyncGenerator[str, None]],
|
||||
) -> T | V | AsyncGenerator[str, None] | ErrorResponse:
|
||||
"""Base method for speech-to-text operations like transcription and
|
||||
translation.
|
||||
"""
|
||||
translation."""
|
||||
if request.stream and request.use_beam_search:
|
||||
return self.create_error_response(
|
||||
"Streaming is not currently supported with beam search"
|
||||
|
||||
Reference in New Issue
Block a user