from __future__ import annotations

from typing import Dict, Iterable, List, Optional, Union

import httpx
from typing_extensions import Literal

from ...._base_client import make_request_options
from ...._compat import cached_property
from ...._resource import AsyncAPIResource, SyncAPIResource
from ...._types import Body, Headers, Query
from ...._utils import async_with_sts_token, with_sts_token
from ...encryption import with_e2e_encryption, async_with_e2e_encryption
from ....types.chat import (
    ChatCompletion,
    ChatCompletionMessageParam,
    ChatCompletionToolChoiceOptionParam,
    ChatCompletionToolParam,
    completion_create_params,
)
from ...chat.completions import (
    AsyncCompletionsWithRawResponse,
    CompletionsWithRawResponse,
)
from .._utils import async_with_batch_retry, get_request_last_time, with_batch_retry

__all__ = ["Completions", "AsyncCompletions"]


class Completions(SyncAPIResource):
    @cached_property
    def with_raw_response(self) -> CompletionsWithRawResponse:
        return CompletionsWithRawResponse(self)

    @with_sts_token
    @with_e2e_encryption
    def create(
        self,
        *,
        messages: Iterable[ChatCompletionMessageParam],
        model: str,
        frequency_penalty: Optional[float] | None = None,
        function_call: completion_create_params.FunctionCall | None = None,
        logit_bias: Optional[Dict[str, int]] | None = None,
        logprobs: Optional[bool] | None = None,
        max_tokens: Optional[int] | None = None,
        presence_penalty: Optional[float] | None = None,
        stop: Union[Optional[str], List[str]] | None = None,
        temperature: Optional[float] | None = None,
        tools: Iterable[ChatCompletionToolParam] | None = None,
        top_logprobs: Optional[int] | None = None,
        top_p: Optional[float] | None = None,
        repetition_penalty: Optional[float] | None = None,
        n: Optional[int] | None = None,
        parallel_tool_calls: Optional[bool] | None = None,
        service_tier: Optional[Literal["auto", "default"]] | None = None,
        tool_choice: ChatCompletionToolChoiceOptionParam | None = None,
        response_format: completion_create_params.ResponseFormat | None = None,
        thinking: completion_create_params.Thinking | None = None,
        max_completion_tokens: Optional[int] | None = None,
        user: str | None = None,
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None = None,
    ) -> ChatCompletion:
        deadline = get_request_last_time(self._client, timeout)
        breaker = self._client.get_model_breaker(model)

        resp = with_batch_retry(
            deadline,
            breaker,
            self._post_without_retry,
            "/batch/chat/completions",
            body={
                "messages": messages,
                "model": model,
                "frequency_penalty": frequency_penalty,
                "function_call": function_call,
                "logit_bias": logit_bias,
                "logprobs": logprobs,
                "max_tokens": max_tokens,
                "presence_penalty": presence_penalty,
                "stop": stop,
                "temperature": temperature,
                "tools": tools,
                "top_logprobs": top_logprobs,
                "top_p": top_p,
                "user": user,
                "repetition_penalty": repetition_penalty,
                "n": n,
                "parallel_tool_calls": parallel_tool_calls,
                "service_tier": service_tier,
                "tool_choice": tool_choice,
                "response_format": response_format,
                "thinking": thinking,
                "max_completion_tokens": max_completion_tokens,
            },
            options=make_request_options(
                extra_headers=extra_headers,
                extra_query=extra_query,
                extra_body=extra_body,
                timeout=timeout,
            ),
            cast_to=ChatCompletion,
        )
        return resp


class AsyncCompletions(AsyncAPIResource):
    @cached_property
    def with_raw_response(self) -> AsyncCompletionsWithRawResponse:
        return AsyncCompletionsWithRawResponse(self)

    @async_with_sts_token
    @async_with_e2e_encryption
    async def create(
        self,
        *,
        messages: Iterable[ChatCompletionMessageParam],
        model: str,
        frequency_penalty: Optional[float] | None = None,
        function_call: completion_create_params.FunctionCall | None = None,
        logit_bias: Optional[Dict[str, int]] | None = None,
        logprobs: Optional[bool] | None = None,
        max_tokens: Optional[int] | None = None,
        presence_penalty: Optional[float] | None = None,
        stop: Union[Optional[str], List[str]] | None = None,
        temperature: Optional[float] | None = None,
        tools: Iterable[ChatCompletionToolParam] | None = None,
        top_logprobs: Optional[int] | None = None,
        top_p: Optional[float] | None = None,
        user: str | None = None,
        repetition_penalty: Optional[float] | None = None,
        n: Optional[int] | None = None,
        parallel_tool_calls: Optional[bool] | None = None,
        service_tier: Optional[Literal["auto", "default"]] | None = None,
        tool_choice: ChatCompletionToolChoiceOptionParam | None = None,
        response_format: completion_create_params.ResponseFormat | None = None,
        thinking: completion_create_params.Thinking | None = None,
        max_completion_tokens: Optional[int] | None = None,
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None = None,
    ) -> ChatCompletion:
        deadline = get_request_last_time(self._client, timeout)
        breaker = await self._client.get_model_breaker(model)

        resp = await async_with_batch_retry(
            deadline,
            breaker,
            self._post_without_retry,
            "/batch/chat/completions",
            body={
                "messages": messages,
                "model": model,
                "frequency_penalty": frequency_penalty,
                "function_call": function_call,
                "logit_bias": logit_bias,
                "logprobs": logprobs,
                "max_tokens": max_tokens,
                "presence_penalty": presence_penalty,
                "stop": stop,
                "temperature": temperature,
                "tools": tools,
                "top_logprobs": top_logprobs,
                "top_p": top_p,
                "user": user,
                "repetition_penalty": repetition_penalty,
                "n": n,
                "parallel_tool_calls": parallel_tool_calls,
                "service_tier": service_tier,
                "tool_choice": tool_choice,
                "response_format": response_format,
                "thinking": thinking,
                "max_completion_tokens": max_completion_tokens,
            },
            options=make_request_options(
                extra_headers=extra_headers,
                extra_query=extra_query,
                extra_body=extra_body,
                timeout=timeout,
            ),
            cast_to=ChatCompletion,
        )

        return resp