from __future__ import annotations

from dataclasses import dataclass
from typing import Literal

Mode = Literal["chat", "work"]
Strength = Literal["light", "medium", "high"]


@dataclass(frozen=True, slots=True)
class ModelSpec:
    public_id: str
    display_name: str
    mode: Mode
    strength: Strength
    provider_model: str
    max_iterations: int
    context_char_budget: int


_TIERS = {
    "light": ("轻度", "ChatGPT-5.6:Luna", 8, 120_000),
    "medium": ("中", "ChatGPT-5.6:Terra", 16, 240_000),
    "high": ("高", "ChatGPT-5.6:Sol", 24, 400_000),
}

MODEL_SPECS: dict[str, ModelSpec] = {
    f"{mode}-{strength}": ModelSpec(
        public_id=f"{mode}-{strength}",
        display_name=f"{'Chat' if mode == 'chat' else 'Work'} · {label}",
        mode=mode,
        strength=strength,
        provider_model=provider,
        max_iterations=max_iterations,
        context_char_budget=context_budget,
    )
    for mode in ("chat", "work")
    for strength, (label, provider, max_iterations, context_budget) in _TIERS.items()
}


def get_model_spec(model_id: str) -> ModelSpec:
    try:
        return MODEL_SPECS[model_id]
    except KeyError as exc:
        raise ValueError(f"Unsupported model: {model_id}") from exc


def openai_model_list() -> dict:
    return {
        "object": "list",
        "data": [
            {
                "id": spec.public_id,
                "object": "model",
                "owned_by": "k1412-agent",
                "name": spec.display_name,
            }
            for spec in MODEL_SPECS.values()
        ],
    }