import argparse
import csv
import os


def parse_arguments():
    parser = argparse.ArgumentParser(
        description="Extract unique taxonomy mappings (FeatureID to taxonomy lineage) from a directory of CSV files."
    )
    parser.add_argument(
        "-i",
        "--input-dir",
        required=True,
        help="Path to the directory containing input CSV files.",
    )
    parser.add_argument(
        "-o",
        "--output",
        required=True,
        help="Path for the output taxonomy CSV file.",
    )
    return parser.parse_args()


def sort_key(val):
    """Sort numeric tax IDs numerically first, followed by string IDs alphabetically."""
    val_str = str(val).strip()
    if val_str.isdigit():
        return (0, int(val_str))
    return (1, val_str)


def extract_taxonomy(input_dir, output_file):
    taxonomy_map = {}
    csv_files = [
        os.path.join(input_dir, f)
        for f in os.listdir(input_dir)
        if f.lower().endswith(".csv")
    ]

    if not csv_files:
        print(f"Error: No CSV files found in '{input_dir}'")
        return

    print(f"Processing {len(csv_files)} CSV file(s)...")

    tax_cols = [
        "Domain",
        "Kingdom",
        "Phylum",
        "Class",
        "Order",
        "Family",
        "Genus",
        "Species",
    ]

    for file_path in csv_files:
        with open(file_path, mode="r", newline="", encoding="utf-8") as infile:
            reader = csv.DictReader(infile)

            for row in reader:
                tax_id = row["taxid_at_target_level"].strip()
                if not tax_id:
                    continue

                # Store the taxonomy annotation for newly encountered feature IDs
                if tax_id not in taxonomy_map:
                    taxonomy_map[tax_id] = {
                        col: row.get(col, "").strip() for col in tax_cols
                    }

    headers = ["FeatureID"] + tax_cols

    with open(output_file, mode="w", newline="", encoding="utf-8") as outfile:
        writer = csv.writer(outfile)
        writer.writerow(headers)

        # Write each unique feature ID and its corresponding taxonomy
        for tax_id in sorted(taxonomy_map.keys(), key=sort_key):
            row = [tax_id] + [taxonomy_map[tax_id][col] for col in tax_cols]
            writer.writerow(row)

    print(
        f"Successfully written {len(taxonomy_map)} unique feature tax ID lineages to '{output_file}'"
    )


def main():
    args = parse_arguments()
    extract_taxonomy(args.input_dir, args.output)


if __name__ == "__main__":
    main()
