scripts/compress_dataset.py

import os
# import tarfile
from pathlib import Path
import argparse

arg_parser = argparse.ArgumentParser()
arg_parser.add_argument("--input", type=str, default=".", help="Path to the folder to compress")
arg_parser.add_argument("--output", type=str, default="EN-SLAM-HF", help="Output Path to the folder to compress")
args = arg_parser.parse_args()

def compress_and_split(path: Path, max_size_gb=2):
    max_size_bytes = max_size_gb * 1024 * 1024 * 1024
    tar_file = f"{args.output}/{path.parent.parent.name}_{path.parent.name}_{path.stem}.tar.gz"
    
    print(f"compressing ... {path}")
    # with tarfile.open(tar_file, "w:gz") as tar: tar.add(path)
    os.system(f"tar -czvf {tar_file} {path}")
    print(f"DONE!")
    
    if os.path.getsize(tar_file) > max_size_bytes:
        print(f"{tar_file} > {max_size_gb}GB, spliting ...")
        split_tar(tar_file, max_size_bytes)
        os.remove(tar_file)
        print(f"FNISHED!")

def split_tar(tar_file, max_size_bytes):
    part_num = 1
    with open(tar_file, "rb") as f:
        while chunk := f.read(max_size_bytes):
            part_file = f"{tar_file}.part{part_num:03d}"
            with open(part_file, "wb") as part:
                part.write(chunk)
            print(f"Generate chunks: {part_file}")
            part_num += 1

def main():
    base_dir = Path(args.input)
    for path in base_dir.iterdir():
        print(f"Processing ... {path}")
        # if path.is_dir() and path.name == "depth":
        compress_and_split(path)


if __name__ == "__main__":
    main()