request.py 2.92 KB
Newer Older
1
# SPDX-License-Identifier: Apache-2.0
2
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
3
4
5
6
7
import dataclasses
import functools
import json
from concurrent.futures import Future
from concurrent.futures._base import TimeoutError
8
from typing import cast
9
10

from vllm.sampling_params import SamplingParams
11
12
13
14
15
from vllm.v1.structured_output.backend_types import (
    StructuredOutputGrammar,
    StructuredOutputKey,
    StructuredOutputOptions,
)
16
17
18
19
20


@dataclasses.dataclass
class StructuredOutputRequest:
    sampling_params: SamplingParams
21
    _grammar: Future[StructuredOutputGrammar] | StructuredOutputGrammar | None = None
22
    reasoning_ended: bool | None = None
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41

    def _check_grammar_completion(self) -> bool:
        # NOTE: We have to lazy import to gate circular imports
        from vllm.v1.request import RequestStatus

        if isinstance(self._grammar, Future):
            try:
                # We will check whether the future is ready within 100 us
                self._grammar = self._grammar.result(timeout=0.0001)
                self.status = RequestStatus.WAITING
            except TimeoutError:
                return False
        return True

    @property
    def is_grammar_ready(self) -> bool:
        return self._check_grammar_completion()

    @property
42
    def grammar(self) -> StructuredOutputGrammar | None:
43
        completed = self._check_grammar_completion()
44
        return (
45
            cast(StructuredOutputGrammar | None, self._grammar) if completed else None
46
        )
47
48

    @grammar.setter
49
    def grammar(
50
        self, grammar: StructuredOutputGrammar | Future[StructuredOutputGrammar]
51
    ) -> None:
52
53
54
55
        self._grammar = grammar

    @functools.cached_property
    def structured_output_key(self) -> StructuredOutputKey:
56
57
58
        return get_structured_output_key(self.sampling_params)


59
def get_structured_output_key(sampling_params: SamplingParams) -> StructuredOutputKey:
60
    params = sampling_params.structured_outputs
61
62
63
64
65
66
67
68
69
70
71
72
73
74
    assert params is not None, "params can't be None."
    if params.json is not None:
        if not isinstance(params.json, str):
            json_str = json.dumps(params.json)
        else:
            json_str = params.json
        return (StructuredOutputOptions.JSON, json_str)
    elif params.json_object:
        return (StructuredOutputOptions.JSON_OBJECT, "")
    elif params.regex is not None:
        return (StructuredOutputOptions.REGEX, params.regex)
    elif params.choice is not None:
        if not isinstance(params.choice, str):
            json_str = json.dumps(params.choice)
75
        else:
76
77
78
79
            json_str = params.choice
        return (StructuredOutputOptions.CHOICE, json_str)
    elif params.grammar is not None:
        return (StructuredOutputOptions.GRAMMAR, params.grammar)
80
81
    elif params.structural_tag is not None:
        return (StructuredOutputOptions.STRUCTURAL_TAG, params.structural_tag)
82
83
    else:
        raise ValueError("No valid structured output parameter found")