"""Fresh-process duplicate-deal comparison against RLCard's original rule bot.
Example: .venvs/doudizhu/bin/python games/doudizhu/tests/compare.py --output /tmp/ddz.json
The new bot occupies each seat once; the other seats use independent rule bots.
"""
import argparse
import json
import pathlib
import sys
import time

import rlcard

ROOT = pathlib.Path(__file__).resolve().parents[1]
sys.path.insert(0, str(ROOT / 'referee'))
from play import player, game

parser = argparse.ArgumentParser()
parser.add_argument('--start', type=int, default=101)
parser.add_argument('--deals', type=int, default=20)
parser.add_argument('--output', type=pathlib.Path)
args = parser.parse_args()
if not 1 <= args.deals <= 100:
    parser.error('--deals must be between 1 and 100')
results, started = [], time.monotonic()
for seed in range(args.start, args.start + args.deals):
    for seat in range(3):
        bots = []
        try:
            for s in range(3):
                bots.append(player(str(ROOT / 'field' / ('strategic' if s == seat else 'rule'))))
            scores, replay = game(rlcard.make('doudizhu'), bots, seed)
            results.append({'seed': seed, 'seat': seat, 'won': scores[seat] > 0,
                            'points': scores[seat], 'fallbacks': sum(replay['fallbacks'])})
        finally:
            for bot in bots:
                bot.close()
    print(json.dumps({'games': len(results), 'wins': sum(r['won'] for r in results)}), flush=True)
summary = {'games': len(results), 'wins': sum(r['won'] for r in results),
           'mean_points': sum(r['points'] for r in results) / len(results),
           'fallbacks': sum(r['fallbacks'] for r in results), 'seconds': time.monotonic() - started}
print(json.dumps(summary))
if args.output:
    args.output.write_text(json.dumps({'summary': summary, 'games': results}, indent=2))