biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Update deprecated Python 3.8 typing (#13971)

Browse Source
This commit is contained in:
Harry Mellor
2025-03-03 01:34:51 +00:00
committed by GitHub
parent bf33700ecd
commit cf069aa8aa
300 changed files with 2294 additions and 2347 deletions
Download Patch File Download Diff File Expand all files Collapse all files

31
vllm/entrypoints/openai/serving_chat.py
View File

@@ -3,10 +3,9 @@
import asyncio
import json
import time
from typing import (AsyncGenerator, AsyncIterator, Callable, Dict, Final, List,
Optional)
from typing import Sequence as GenericSequence
from typing import Union
from collections.abc import AsyncGenerator, AsyncIterator
from collections.abc import Sequence as GenericSequence
from typing import Callable, Final, Optional, Union
from fastapi import Request
@@ -205,7 +204,7 @@ class OpenAIServingChat(OpenAIServing):
raw_request.state.request_metadata = request_metadata
# Schedule the request and get the result generator.
generators: List[AsyncGenerator[RequestOutput, None]] = []
generators: list[AsyncGenerator[RequestOutput, None]] = []
try:
for i, engine_prompt in enumerate(engine_prompts):
sampling_params: Union[SamplingParams, BeamSearchParams]
@@ -282,7 +281,7 @@ class OpenAIServingChat(OpenAIServing):
result_generator: AsyncIterator[RequestOutput],
request_id: str,
model_name: str,
conversation: List[ConversationMessage],
conversation: list[ConversationMessage],
tokenizer: AnyTokenizer,
request_metadata: RequestResponseMetadata,
) -> AsyncGenerator[str, None]:
@@ -310,7 +309,7 @@ class OpenAIServingChat(OpenAIServing):
should_stream_with_reasoning_parsing = (
self._should_stream_with_reasoning_parsing(request))
all_previous_token_ids: Optional[List[List[int]]]
all_previous_token_ids: Optional[list[list[int]]]
# Only one of these will be used, thus previous_texts and
# all_previous_token_ids will not be used twice in the same iteration.
@@ -339,7 +338,7 @@ class OpenAIServingChat(OpenAIServing):
# Prepare the tool parser if it's needed
try:
if tool_choice_auto and self.tool_parser:
tool_parsers: List[Optional[ToolParser]] = [
tool_parsers: list[Optional[ToolParser]] = [
self.tool_parser(tokenizer)
] * num_choices
else:
@@ -406,7 +405,7 @@ class OpenAIServingChat(OpenAIServing):
# Send response to echo the input portion of the
# last message
if request.echo:
last_msg_content: Union[str, List[Dict[str, str]]] = ""
last_msg_content: Union[str, list[dict[str, str]]] = ""
if conversation and "content" in conversation[
-1] and conversation[-1].get("role") == role:
last_msg_content = conversation[-1]["content"] or ""
@@ -674,7 +673,7 @@ class OpenAIServingChat(OpenAIServing):
result_generator: AsyncIterator[RequestOutput],
request_id: str,
model_name: str,
conversation: List[ConversationMessage],
conversation: list[ConversationMessage],
tokenizer: AnyTokenizer,
request_metadata: RequestResponseMetadata,
) -> Union[ErrorResponse, ChatCompletionResponse]:
@@ -693,7 +692,7 @@ class OpenAIServingChat(OpenAIServing):
assert final_res is not None
choices: List[ChatCompletionResponseChoice] = []
choices: list[ChatCompletionResponseChoice] = []
role = self.get_chat_request_role(request)
for output in final_res.outputs:
@@ -812,7 +811,7 @@ class OpenAIServingChat(OpenAIServing):
choices.append(choice_data)
if request.echo:
last_msg_content: Union[str, List[Dict[str, str]]] = ""
last_msg_content: Union[str, list[dict[str, str]]] = ""
if conversation and "content" in conversation[-1] and conversation[
-1].get("role") == role:
last_msg_content = conversation[-1]["content"] or ""
@@ -853,8 +852,8 @@ class OpenAIServingChat(OpenAIServing):
return response
def _get_top_logprobs(
self, logprobs: Dict[int, Logprob], top_logprobs: Optional[int],
tokenizer: AnyTokenizer) -> List[ChatCompletionLogProb]:
self, logprobs: dict[int, Logprob], top_logprobs: Optional[int],
tokenizer: AnyTokenizer) -> list[ChatCompletionLogProb]:
return [
ChatCompletionLogProb(token=(token := self._get_decoded_token(
p[1],
@@ -871,12 +870,12 @@ class OpenAIServingChat(OpenAIServing):
def _create_chat_logprobs(
self,
token_ids: GenericSequence[int],
top_logprobs: GenericSequence[Optional[Dict[int, Logprob]]],
top_logprobs: GenericSequence[Optional[dict[int, Logprob]]],
tokenizer: AnyTokenizer,
num_output_top_logprobs: Optional[int] = None,
) -> ChatCompletionLogProbs:
"""Create OpenAI-style logprobs."""
logprobs_content: List[ChatCompletionLogProbsContent] = []
logprobs_content: list[ChatCompletionLogProbsContent] = []
for i, token_id in enumerate(token_ids):
step_top_logprobs = top_logprobs[i]