from argparse import ArgumentParser
from glob import glob
from typing import List, Union

from pandas import DataFrame, read_html


def find_html_files(path) -> List:
    file_list = glob(path + "/*fastqc.html")
    return file_list


def extract_adapters(files) -> Union[DataFrame, None]:
    all_adapters = DataFrame()
    for entry in files:
        tables = read_html(entry)
        adapter_sequences = tables[1].Sequence
        all_adapters = all_adapters.append(adapter_sequences)
    return all_adapters.dropna()


def save_to_file(filename, adapters) -> None:
    adapters.to_csv(filename, index=False, header=False, sep="\n")


def parse_arguments():
    parser = ArgumentParser()
    parser.add_argument("input", help="directory containing the fastqc reports")
    parser.add_argument("output", help="file where to export the sequences")
    return parser.parse_args()


def main():
    args = parse_arguments()
    file_list = find_html_files(args.input)
    adapters = extract_adapters(file_list)
    save_to_file(args.output, adapters)


if __name__ == "__main__":
    main()