import argparse
import sys
import csv
#import lib.sort

import os

def default_sort(rows, top_score_metric):
    sorted_rows = sorted(rows, key=lambda row: (int(row['Sub-peptide Position'])))
    sorted_rows = sorted(sorted_rows, key=lambda row: (float(row['Corresponding Fold Change']) if row['Corresponding Fold Change'].isdigit() else float('inf')), reverse=True)
    if top_score_metric == 'median':
        sorted_rows = sorted(
            sorted_rows,
            key=lambda row: (
                row['Gene Name'],
                row['Mutation'],
                float(row['Median MT Score']),
            )
        )
    elif top_score_metric == 'lowest':
        sorted_rows = sorted(
            sorted_rows,
            key=lambda row: (
                row['Gene Name'],
                row['Mutation'],
                float(row['Best MT Score']),
            )
        )
    return sorted_rows


def main(input_files, output_file, top_score_metric):
    output_file = open(output_file, 'w')

    fieldnames = []
    for input_file in input_files:
        with open(input_file, 'r') as input_file_handle:
            reader = csv.DictReader(input_file_handle, delimiter='\t')
            if len(fieldnames) == 0:
                fieldnames = reader.fieldnames
            else:
                for fieldname in reader.fieldnames:
                    if fieldname not in fieldnames:
                        fieldnames.append(fieldname)

    rows = []
    for input_file in input_files:
        with open(input_file, 'r') as input_file_handle:
            reader = csv.DictReader(input_file_handle, delimiter='\t')
            for row in reader:
                for fieldname in fieldnames:
                    if fieldname not in row:
                        row[fieldname] = 'NA'
                rows.append(row)

    sorted_rows = default_sort(rows, top_score_metric)
    tsv_writer = csv.DictWriter(output_file, list(fieldnames), delimiter = '\t', lineterminator = '\n')
    tsv_writer.writeheader()
    tsv_writer.writerows(sorted_rows)

    output_file.close()


if __name__ == "__main__":
    files = []
    for file in os.listdir("/home/ben/Documents/wustl/pvacdata/H_NT-45-1208990-indels/MHC_Class_I/tmp"):
        if ".parsed.tsv" in file:
            files.append(os.path.join("/home/ben/Documents/wustl/pvacdata/H_NT-45-1208990-indels/MHC_Class_I/tmp", file))
    main(files, "/home/ben/Documents/wustl/pvacdata/H_NT-45-1208990-indels/H_NT-45-1208990.all_epitopes.tsv ", "median")