from __future__ import annotations

import os
from functools import lru_cache
from typing import Literal

from pydantic import BaseModel, Field

Provider = Literal["openai", "anthropic"]


class LLMSettings(BaseModel):
    provider: Provider = Field(default=os.getenv("LLM_PROVIDER", "openai"))
    api_key: str = Field(default=os.getenv("LLM_API_KEY", "none"))
    model: str = Field(default=os.getenv("LLM_MODEL", "qwen2.5-7b-instruct.q4_k_m"))
    base_url: str = Field(default=os.getenv("LLM_BASE_URL", "http://ollama.internal.henryhosted.com:9292/v1"))
    temperature: float = Field(default=float(os.getenv("LLM_TEMPERATURE", "0.1")))
    timeout_seconds: float = Field(default=float(os.getenv("LLM_TIMEOUT_SECONDS", "60")))
    max_retries: int = Field(default=int(os.getenv("LLM_MAX_RETRIES", "3")))


@lru_cache(maxsize=1)
def get_settings() -> LLMSettings:
    return LLMSettings()


def get_request_settings(
    provider: str | None = None,
    model: str | None = None,
    base_url: str | None = None,
    api_key: str | None = None,
    temperature: float | None = None,
) -> LLMSettings:
    base = get_settings()
    data = base.model_dump()
    if provider is not None:
        data["provider"] = provider
    if model is not None:
        data["model"] = model
    if base_url is not None:
        data["base_url"] = base_url
    if api_key is not None:
        data["api_key"] = api_key
    if temperature is not None:
        data["temperature"] = temperature
    return LLMSettings(**data)