"""Portable local EmbodiedTrace loader. Python standard library only."""
import json
from pathlib import Path
from zipfile import ZipFile
SPLITS=('train','validation','test')
def iter_decisions(root, split='train'):
    if split not in SPLITS:raise ValueError(f'split must be one of {SPLITS}')
    with (Path(root)/'data'/f'{split}.jsonl').open(encoding='utf-8') as source:
        for line in source:
            if line.strip():yield json.loads(line)
def extract_media(root, kind='images'):
    if kind not in ('images','videos'):raise ValueError('kind must be images or videos')
    root=Path(root).resolve()
    with ZipFile(root/'media'/f'{kind}.zip') as archive:
        for name in archive.namelist():
            target=(root/name).resolve()
            if not target.is_relative_to(root):raise ValueError('Unsafe archive path')
        archive.extractall(root)
    return root/kind
if __name__=='__main__':
    import argparse
    parser=argparse.ArgumentParser();parser.add_argument('root');parser.add_argument('--split',default='train');args=parser.parse_args()
    row=next(iter_decisions(args.root,args.split));print(json.dumps({'episode':row['episode_id'],'subgoal':row['generated_annotation']['subgoal']['title'],'action':row['action']},indent=2))