#!/usr/bin/env python3
"""
benchmark/camera_count/analyze.py
===================================
Liest camera_study.json und erstellt:
  - Konsolentabelle: k → Mittelwert / Median / Std / Min / Max des Fehlers
    (inkl. n/a-Spalte: Subsets, bei denen der Punkt unbeobachtbar war)
  - Beste und schlechteste Kamera-Subsets je k
  - Boxplot: Anzahl Kameras vs. Fehler (PNG)

Metriken:
  finger_error_mm (Standard) — Fingerposition in mm (volle Kette x..e).
                                Leer, wenn Hand/Palm/Finger unbeobachtbar.
  wrist_error_mm             — Handgelenkposition in mm (nur Armgelenke x,y,z,a).
  mean_abs_deg / max_abs_deg — Gelenkwinkelfehler in Grad (nur beobachtbare Gelenke).
  mean_abs_mm / max_abs_mm   — Lineargelenkfehler (x, e) in mm.

Aufruf:
  python benchmark/camera_count/analyze.py
  python benchmark/camera_count/analyze.py --metric wrist_error_mm
  python benchmark/camera_count/analyze.py --metric mean_abs_deg
"""
from __future__ import annotations

import argparse
import json
import statistics
from pathlib import Path

RESULTS_DIR = Path(__file__).resolve().parent / "results"


def load(path: str) -> list[dict]:
    data = json.loads(Path(path).read_text(encoding="utf-8"))
    if not data:
        print("[ERROR] Keine Ergebnisse in der Datei.")
    return data


def group_by_k(data: list[dict], metric: str) -> tuple[dict[int, list[float]], dict[int, int]]:
    """Werte je k (None übersprungen) plus Anzahl der n/a-Fälle je k."""
    by_k: dict[int, list[float]] = {}
    na_by_k: dict[int, int] = {}
    for row in data:
        k = row["k"]
        v = row.get(metric)
        if v is None:
            na_by_k[k] = na_by_k.get(k, 0) + 1
            continue
        by_k.setdefault(k, []).append(v)
    return by_k, na_by_k


def print_table(by_k: dict[int, list[float]], na_by_k: dict[int, int], metric: str) -> None:
    print(f"\nKamera-Anzahl vs. {metric}")
    print(f"{'k':>4} | {'n':>5} | {'n/a':>5} | {'Mittel':>8} | {'Median':>8} | "
          f"{'Std':>8} | {'Min':>8} | {'Max':>8}")
    print("-" * 74)
    all_ks = sorted(set(by_k) | set(na_by_k))
    for k in all_ks:
        na = na_by_k.get(k, 0)
        vals = by_k.get(k, [])
        if not vals:
            print(f"{k:>4} | {0:>5} | {na:>5} | {'—':>8} | {'—':>8} | "
                  f"{'—':>8} | {'—':>8} | {'—':>8}")
            continue
        print(f"{k:>4} | {len(vals):>5} | {na:>5} | "
              f"{statistics.mean(vals):8.3f} | {statistics.median(vals):8.3f} | "
              f"{statistics.pstdev(vals):8.3f} | {min(vals):8.3f} | {max(vals):8.3f}")
    if any(na_by_k.values()):
        print("\n  n/a = Subsets, bei denen dieser Punkt unbeobachtbar war "
              "(Position unbekannt, nicht eingerechnet).")


def print_best_worst(data: list[dict], metric: str) -> None:
    ks = sorted({r["k"] for r in data})
    print(f"\nBeste / schlechteste Subsets je k  ({metric}):")
    for k in ks:
        rows_k = [r for r in data if r["k"] == k and r.get(metric) is not None]
        if not rows_k:
            print(f"  k={k}:  keine beobachtbaren Werte")
            continue
        best = min(rows_k, key=lambda r: r[metric])
        worst = max(rows_k, key=lambda r: r[metric])
        print(f"  k={k}:  best  [{best['scene']}] {best['subset']}  "
              f"({best[metric]:.3f})   "
              f"worst [{worst['scene']}] {worst['subset']}  ({worst[metric]:.3f})")


def plot(by_k: dict[int, list[float]], na_by_k: dict[int, int],
         metric: str, out_path: str) -> None:
    try:
        import matplotlib.pyplot as plt
    except ImportError:
        print("\n[INFO] matplotlib nicht installiert — Plot übersprungen.")
        print("       pip install matplotlib")
        return

    ks = sorted(by_k)
    if not ks:
        print("\n[INFO] Keine Werte zum Plotten (alle unbeobachtbar).")
        return
    data_plot = [by_k[k] for k in ks]

    fig, ax = plt.subplots(figsize=(8, 5))
    ax.boxplot(data_plot, labels=[str(k) for k in ks], patch_artist=True)
    ax.set_xlabel("Anzahl Kameras")
    ax.set_ylabel(metric)
    ax.set_title("Anzahl Kameras vs. Pose-Schätzfehler")
    ax.grid(True, axis="y", alpha=0.35)

    ymax = max(max(v) for v in data_plot)
    for i, k in enumerate(ks, start=1):
        na = na_by_k.get(k, 0)
        note = f"n={len(by_k[k])}" + (f"\nn/a={na}" if na else "")
        ax.text(i, ymax * 1.02, note, ha="center", va="bottom", fontsize=8, color="gray")

    fig.tight_layout()
    fig.savefig(out_path, dpi=150)
    print(f"\n[INFO] Plot gespeichert: {out_path}")


def main() -> None:
    ap = argparse.ArgumentParser(description="Auswertung der Kamera-Anzahl-Studie")
    ap.add_argument("--input", default=str(RESULTS_DIR / "camera_study.json"))
    ap.add_argument("--metric",
                    choices=["finger_error_mm", "wrist_error_mm",
                             "mean_abs_deg", "max_abs_deg", "mean_abs_mm", "max_abs_mm"],
                    default="finger_error_mm",
                    help="Metrik für Tabelle und Plot (Standard: finger_error_mm)")
    ap.add_argument("--out-plot", default=None,
                    help="Plot-Pfad (Standard: results/camera_count_<metric>.png)")
    args = ap.parse_args()

    data = load(args.input)
    if not data:
        return

    by_k, na_by_k = group_by_k(data, args.metric)
    if not by_k and not na_by_k:
        print(f"[ERROR] Keine Werte für Metrik '{args.metric}' gefunden.")
        return

    out_plot = args.out_plot or str(RESULTS_DIR / f"camera_count_{args.metric}.png")

    print_table(by_k, na_by_k, args.metric)
    print_best_worst(data, args.metric)
    plot(by_k, na_by_k, args.metric, out_plot)


if __name__ == "__main__":
    main()