Fix per file ruff ignores related to typing (#26254)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-10-05 16:37:55 +00:00
committed by GitHub
parent 5f317530ec
commit 1c0c68202c
32 changed files with 258 additions and 285 deletions
+9 -9
View File
@@ -4,7 +4,7 @@
from __future__ import annotations
from dataclasses import dataclass
from typing import TYPE_CHECKING, Optional
from typing import TYPE_CHECKING
from vllm._bc_linter import bc_linter_include
@@ -25,14 +25,14 @@ if TYPE_CHECKING:
@dataclass
class NewRequestData:
req_id: str
prompt_token_ids: Optional[list[int]]
prompt_token_ids: list[int] | None
mm_features: list[MultiModalFeatureSpec]
sampling_params: Optional[SamplingParams]
pooling_params: Optional[PoolingParams]
sampling_params: SamplingParams | None
pooling_params: PoolingParams | None
block_ids: tuple[list[int], ...]
num_computed_tokens: int
lora_request: Optional[LoRARequest]
prompt_embeds: Optional[torch.Tensor] = None
lora_request: LoRARequest | None
prompt_embeds: torch.Tensor | None = None
@classmethod
def from_request(
@@ -98,7 +98,7 @@ class CachedRequestData:
# NOTE(woosuk): new_token_ids is only used for pipeline parallelism.
# When PP is not used, new_token_ids will be empty.
new_token_ids: list[list[int]]
new_block_ids: list[Optional[tuple[list[int], ...]]]
new_block_ids: list[tuple[list[int], ...] | None]
num_computed_tokens: list[int]
num_output_tokens: list[int]
@@ -160,7 +160,7 @@ class SchedulerOutput:
# for filling the next token bitmask
structured_output_request_ids: dict[str, int]
# the bitmask for the whole batch
grammar_bitmask: Optional[npt.NDArray[np.int32]]
grammar_bitmask: npt.NDArray[np.int32] | None
# KV Cache Connector metadata.
kv_connector_metadata: Optional[KVConnectorMetadata] = None
kv_connector_metadata: KVConnectorMetadata | None = None
+12 -12
View File
@@ -7,7 +7,7 @@ import itertools
import time
from collections import defaultdict
from collections.abc import Iterable
from typing import Any, Optional, Union
from typing import Any, Union
from vllm.config import VllmConfig
from vllm.distributed.kv_events import EventPublisherFactory, KVEventBatch
@@ -64,7 +64,7 @@ class Scheduler(SchedulerInterface):
# request ids should be included in the EngineCoreOutputs returned
# by update_from_outputs(). This is currently used in the multi-engine
# case to track request lifetimes efficiently.
self.finished_req_ids_dict: Optional[dict[int, set[str]]] = (
self.finished_req_ids_dict: dict[int, set[str]] | None = (
defaultdict(set) if include_finished_set else None
)
@@ -708,7 +708,7 @@ class Scheduler(SchedulerInterface):
) -> CachedRequestData:
req_ids: list[str] = []
new_token_ids: list[list[int]] = []
new_block_ids: list[Optional[tuple[list[int], ...]]] = []
new_block_ids: list[tuple[list[int], ...] | None] = []
num_computed_tokens: list[int] = []
num_output_tokens: list[int] = []
@@ -921,7 +921,7 @@ class Scheduler(SchedulerInterface):
kv_connector_output = model_runner_output.kv_connector_output
outputs: dict[int, list[EngineCoreOutput]] = defaultdict(list)
spec_decoding_stats: Optional[SpecDecodingStats] = None
spec_decoding_stats: SpecDecodingStats | None = None
kv_connector_stats = (
kv_connector_output.kv_connector_stats if kv_connector_output else None
)
@@ -1212,7 +1212,7 @@ class Scheduler(SchedulerInterface):
request.status = finished_status
self._free_request(request)
def _free_request(self, request: Request) -> Optional[dict[str, Any]]:
def _free_request(self, request: Request) -> dict[str, Any] | None:
assert request.is_finished()
delay_free_blocks, kv_xfer_params = self._connector_finished(request)
@@ -1243,9 +1243,9 @@ class Scheduler(SchedulerInterface):
def make_stats(
self,
spec_decoding_stats: Optional[SpecDecodingStats] = None,
kv_connector_stats: Optional[KVConnectorStats] = None,
) -> Optional[SchedulerStats]:
spec_decoding_stats: SpecDecodingStats | None = None,
kv_connector_stats: KVConnectorStats | None = None,
) -> SchedulerStats | None:
if not self.log_stats:
return None
prefix_cache_stats = self.kv_cache_manager.make_prefix_cache_stats()
@@ -1262,10 +1262,10 @@ class Scheduler(SchedulerInterface):
def make_spec_decoding_stats(
self,
spec_decoding_stats: Optional[SpecDecodingStats],
spec_decoding_stats: SpecDecodingStats | None,
num_draft_tokens: int,
num_accepted_tokens: int,
) -> Optional[SpecDecodingStats]:
) -> SpecDecodingStats | None:
if not self.log_stats:
return None
if spec_decoding_stats is None:
@@ -1285,12 +1285,12 @@ class Scheduler(SchedulerInterface):
# KV Connector Related Methods
########################################################################
def get_kv_connector(self) -> Optional[KVConnectorBase_V1]:
def get_kv_connector(self) -> KVConnectorBase_V1 | None:
return self.connector
def _connector_finished(
self, request: Request
) -> tuple[bool, Optional[dict[str, Any]]]:
) -> tuple[bool, dict[str, Any] | None]:
"""
Invoke the KV connector request_finished() method if applicable.