import os
from dataclasses import dataclass, field
from pathlib import Path

from dotenv import load_dotenv


@dataclass
class TranscriberConfig:
    """Configuration for the transcription pipeline."""

    input_path: str = ""
    output_dir: str = "./output"
    model: str = "large-v3"
    device: str = "cuda"
    compute_type: str = "float16"
    language: str = "ru"
    beam_size: int = 5
    vad: bool = True
    max_speakers: int | None = None
    min_speakers: int | None = None
    formats: list[str] = field(default_factory=lambda: ["txt", "json"])
    pause_threshold: float = 1.5
    chunk_duration: int = 1800
    hf_token: str = ""

    def __post_init__(self):
        load_dotenv()
        if not self.hf_token:
            self.hf_token = os.getenv("HF_TOKEN", "")
        if not self.hf_token:
            raise ValueError(
                "HF_TOKEN is required for pyannote diarization. "
                "Set it in .env or pass via --hf-token"
            )
        if self.device == "cpu":
            self.compute_type = "int8"
        Path(self.output_dir).mkdir(parents=True, exist_ok=True)