from __future__ import annotations

import json
import logging
from pathlib import Path

from umbraaudio.analysis.voice_analysis import VoiceAnalysisEngine
from umbraaudio.config import load_config

logger = logging.getLogger(__name__)


def run_voice_analysis(project_dir: Path | None = None) -> None:
    """Compute the objective voice-analysis report for a project."""
    project_dir = project_dir or Path(".")
    cfg = load_config(project_dir)

    speakers_path = cfg.data_dir / "speakers.json"
    speaker_names: set[str] = set()
    if speakers_path.exists():
        with open(speakers_path) as f:
            speaker_names = {s["primary_name"] for s in json.load(f).get("speakers", [])}

    engine = VoiceAnalysisEngine(
        refs_dir=cfg.data_dir / "voices",
        output_dir=cfg.data_dir / "analysis",
        target_sample_rate=cfg.tts.sample_rate,
        speaker_names=speaker_names,
    )
    report = engine.analyze()
    summary = report["summary"]
    print(
        f"Analyzed {summary['total']} voices: "
        f"{summary['passed']} passed, {summary['failed']} failed, "
        f"{summary['near_duplicate_count']} near-duplicate pairs"
    )
    if report["near_duplicates"]:
        print("Near-duplicate pairs:")
        for pair in report["near_duplicates"]:
            print(f"  {pair['a']} <-> {pair['b']} ({pair['similarity']})")
    if report["errors"]:
        print("Errors:")
        for name, err in report["errors"].items():
            print(f"  {name}: {err}")
    print(f"Report written to {engine.report_path}")


if __name__ == "__main__":
    import sys

    logging.basicConfig(level=logging.INFO, format="%(levelname)s %(message)s")
    project = Path(sys.argv[1]) if len(sys.argv) > 1 else None
    run_voice_analysis(project)