[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
@@ -871,8 +871,7 @@ class AsyncLLM(EngineClient):
|
||||
) -> None:
|
||||
"""Submit a pre-aborted request so the connector's request_finished
|
||||
hook runs to free any pre-admission KV-transfer resources (e.g. NIXL
|
||||
prefill blocks pinned on the P node).
|
||||
"""
|
||||
prefill blocks pinned on the P node)."""
|
||||
request = EngineCoreRequest(
|
||||
request_id=request_id,
|
||||
prompt_token_ids=[0],
|
||||
|
||||
@@ -598,8 +598,7 @@ class EngineCore:
|
||||
|
||||
def _should_throttle_prefills(self) -> bool:
|
||||
"""Whether to defer new prefills this step (DP prefill balancing).
|
||||
Overridden by the DP engine core; never throttles otherwise.
|
||||
"""
|
||||
Overridden by the DP engine core; never throttles otherwise."""
|
||||
return False
|
||||
|
||||
def step(self) -> tuple[dict[int, EngineCoreOutputs], bool]:
|
||||
@@ -1623,8 +1622,7 @@ class EngineCoreProc(EngineCore):
|
||||
@staticmethod
|
||||
def _convert_msgspec_args(method, args):
|
||||
"""If a provided arg type doesn't match corresponding target method
|
||||
arg type, try converting to msgspec object.
|
||||
"""
|
||||
arg type, try converting to msgspec object."""
|
||||
if not args:
|
||||
return args
|
||||
arg_types = signature(method).parameters.values()
|
||||
@@ -2006,8 +2004,7 @@ class EngineCoreProc(EngineCore):
|
||||
|
||||
class DPEngineCoreProc(EngineCoreProc):
|
||||
"""ZMQ-wrapper for running EngineCore in background process
|
||||
in a data parallel context.
|
||||
"""
|
||||
in a data parallel context."""
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
|
||||
@@ -248,8 +248,7 @@ class EngineCoreClient(ABC):
|
||||
|
||||
def dp_engines_running(self) -> bool:
|
||||
"""Returns True if data parallel engines are collectively in a
|
||||
running state.
|
||||
"""
|
||||
running state."""
|
||||
raise NotImplementedError
|
||||
|
||||
async def commit_elastic_ep(self) -> None:
|
||||
@@ -446,8 +445,7 @@ class InprocClient(EngineCoreClient):
|
||||
@dataclass
|
||||
class BackgroundResources:
|
||||
"""Used as a finalizer for clean shutdown, avoiding
|
||||
circular reference back to the client object.
|
||||
"""
|
||||
circular reference back to the client object."""
|
||||
|
||||
ctx: zmq.Context
|
||||
# If CoreEngineProcManager, it manages local engines;
|
||||
@@ -801,8 +799,7 @@ class MPClient(EngineCoreClient):
|
||||
|
||||
def _apply_ready_response(self, payload: bytes) -> None:
|
||||
"""Decode an EngineCoreReadyResponse and sync any post-initialization
|
||||
config changes (e.g. auto-fitted max_model_len) back to the frontend.
|
||||
"""
|
||||
config changes (e.g. auto-fitted max_model_len) back to the frontend."""
|
||||
vllm_config = self.vllm_config
|
||||
response = (
|
||||
msgspec.msgpack.decode(payload, type=EngineCoreReadyResponse)
|
||||
@@ -1333,8 +1330,7 @@ class AsyncMPClient(MPClient):
|
||||
|
||||
class DPAsyncMPClient(AsyncMPClient):
|
||||
"""Asyncio-compatible client for multi-proc, multi-engine (data parallel)
|
||||
EngineCore. Assumes external load-balancing by default.
|
||||
"""
|
||||
EngineCore. Assumes external load-balancing by default."""
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
@@ -1518,8 +1514,7 @@ class DPAsyncMPClient(AsyncMPClient):
|
||||
|
||||
class DPLBAsyncMPClient(DPAsyncMPClient):
|
||||
"""Asyncio-compatible client for multi-proc, multi-engine (data parallel)
|
||||
EngineCore. Load-balances between multiple engine processes.
|
||||
"""
|
||||
EngineCore. Load-balances between multiple engine processes."""
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
@@ -1827,8 +1822,7 @@ class DPLBAsyncMPClient(DPAsyncMPClient):
|
||||
num_redundant_experts: int,
|
||||
) -> None:
|
||||
"""Prepare scale up by creating new engine cores and reconfiguring
|
||||
existing ones.
|
||||
"""
|
||||
existing ones."""
|
||||
self._setup_elastic_ep_reconfig_bootstrap()
|
||||
|
||||
# Phase 1: Send reconfig messages to existing engines
|
||||
@@ -1920,8 +1914,7 @@ class DPLBAsyncMPClient(DPAsyncMPClient):
|
||||
|
||||
async def _commit_scale_down_elastic_ep(self, new_data_parallel_size: int) -> None:
|
||||
"""Scale down the data parallel size by shutting down and
|
||||
reconfiguring existing engine cores.
|
||||
"""
|
||||
reconfiguring existing engine cores."""
|
||||
cur_data_parallel_size = len(self.core_engines)
|
||||
|
||||
self.eep_scaling_cache = ElasticScalingCache(
|
||||
|
||||
@@ -147,8 +147,7 @@ class BaseIncrementalDetokenizer(IncrementalDetokenizer, ABC):
|
||||
|
||||
def get_next_output_text(self, finished: bool, delta: bool) -> str:
|
||||
"""If delta is True, only new text since the last call to
|
||||
this method is returned
|
||||
"""
|
||||
this method is returned"""
|
||||
# We return the full output text if the sequence is finished.
|
||||
buffer_length = 0 if finished else self.stop_buffer_length
|
||||
if not delta:
|
||||
|
||||
@@ -1041,8 +1041,7 @@ def get_engine_zmq_addresses(
|
||||
|
||||
Set ``defer_api_server_ports=False`` only when the consumer cannot
|
||||
report a bound port back (e.g. the Rust front-end). IPC paths are
|
||||
unaffected.
|
||||
"""
|
||||
unaffected."""
|
||||
parallel_config = vllm_config.parallel_config
|
||||
local_engine_count = parallel_config.data_parallel_size_local
|
||||
local_start_index = parallel_config.data_parallel_rank_local
|
||||
|
||||
Reference in New Issue
Block a user