cli.py

#!/usr/bin/env python

import argparse
import re

from convert import convert

parser = argparse.ArgumentParser()
parser.add_argument("input", help="Path to read from")
parser.add_argument(
    "output",
    default=None,
    nargs="?",
    help='Path to write to; defaults to "%s.zarr" % <input>',
)
parser.add_argument(
    "--genome",
    "-g",
    dest="genome",
    required=False,
    help="Top-level 'genome' dataset in a 10x HDF5 (.h5) file",
)
parser.add_argument(
    "--force",
    "-f",
    dest="overwrite",
    type=bool,
    default=False,
    required=False,
    help="If true, overwrite any existing output file that exists (if False, raise an Exception in this case)",
)
parser.add_argument(
    "--chunks",
    "-c",
    dest="chunks",
    required=False,
    default="16m",
    help="For zarr outputs, chunk sizes for 2d arrays",
)


args = parser.parse_args()


def bytes_value(s):
    scales = {"": 1, "k": 1024, "m": 1024 * 1024, "g": 1024 * 1024 * 1024}

    m = re.match("(\d+)(?:([kmg]?)b?)", s.lower())

    if not m:
        raise Exception("Invalid bytes string: %s" % s)

    return int(m.group(1)) * scales[m.group(2)]


convert(
    args.input,
    args.output if args.output else (args.input + ".zarr"),
    genome=args.genome,
    chunk_size=bytes_value(args.chunks),
    overwrite=args.overwrite,
)