Command line utility for encoding and decoding SEQC 3-bit representations of DNA

dna3bit.py

import argparse
import sys
from functools import reduce
from typing import Iterator

def encode(raw: str) -> int:
    return reduce(lambda x, y: x * 8 + y,
        map(lambda x: x % 32 % 5 + 3, raw.encode()))

def _decode(cooked: int) -> Iterator[int]:
    temporary = cooked
    while temporary > 0:
        yield (84, 65, 71, 67, 78)[(temporary & 7) - 3]
        temporary //= 8

def decode(cooked: int) -> str:
    return bytes(_decode(cooked))[::-1].decode()

def main() -> int:
    parser = argparse.ArgumentParser()
    parser.add_argument('-s', '--separator', default=',')
    parser.add_argument('-e', '--encode', nargs='+', type=int, default=[])
    parser.add_argument('-d', '--decode', nargs='+', type=int, default=[])
    args = parser.parse_args()
    for line in sys.stdin:
        columns = line.split(args.separator)
        if len(args.encode) == 0 and len(args.decode) == 0:
            columns = [decode(int(column.strip())) for column in columns]
        else:
            for idx in args.encode:
                columns[idx] = str(encode(columns[idx].strip()))
            for idx in args.decode:
                columns[idx] = decode(int(columns[idx].strip()))
        print(args.separator.join(columns))
    return 0

if __name__ == '__main__':
    sys.exit(main())