Convert SSDI to CSV

convert.py

import csv
from dataclasses import field


key = {
    "0":[[0],'(A)dd/(C)hange/(D)elete'],
    "1":[[2,11], 'SS Number'],
    "2":[[10,31], 'Last name'],
    "3":[[30,35], 'Name Suffix'],
    "4":[[34,50], 'First Name'],
    "5":[[49,65], 'Middle Name'],
    "6":[[64],'(V)erified/(P)roof'],
    "7":[[65,73], 'Date of Death (mmddccyy)'],
    "8":[[73,82], 'Date of Birth (mmddccyy)'],
    "9":[[81,84], 'Blanks'],
    "10":[[83,89], 'Blanks'],
    "11":[[88,94], 'Blanks'],
    "12":[[93,101], 'Blanks']}

def split_using_key(line):
    clean_data = []
    for k,v in key.items():
        indxs, title = v
        if k == '7' or k == '8':
            # parse date
            v, v2 = indxs
            date = line[v:v2]
            mm, dd, yy = [date[0:2], date[2:4], date[4:]]
            clean_data.append("/".join([mm,dd,yy]))
        elif len(indxs) == 1:
            v = indxs[0]
            clean_data.append(line[v])
        elif len(indxs) == 2:
            v, v2 = indxs
            clean_data.append(line[v:v2])
    return clean_data

FILENAME = 'ssdm1'

with open(f'{FILENAME}_processed.csv', 'w', encoding='utf-8') as outfile:
    fields = [a[1] for a in key.values()]
    writer = csv.DictWriter(outfile, fieldnames=fields)
    writer.writeheader()
    with open(FILENAME, 'r',encoding='utf-8') as f:
        while True:
            chunk = f.readlines(5000)
            for line in chunk:
                values = split_using_key(line)
                writer.writerow({fields[count]:values[count] for count,_ in enumerate(values)})