Naturalisation decree parser

from sys import argv
from os import listdir
from os.path import isfile, join
from tika import parser
import re

mentioned_decrees = set()
decrees_dict = {}

parse_dir = argv[1]
min_year = 0 if len(argv) < 3 else argv[2]
files = sorted([f for f in listdir(parse_dir) if f.endswith('.pdf') and isfile(join(parse_dir, f))])

for f in files:
    decree_date = f.split('_')[1]
    decree_date = f'{decree_date[:4]}/{decree_date[4:6]}/{decree_date[6:]}'
    text = parser\
        .from_file(join(parse_dir, f))['content']\
        .replace(' ', '')\
        .replace('\n', '')\
        .replace('\r', '')\
        .replace('\t', '')
    decree_number_regex = re.compile(r'20\d\dX\d{6}')
    decrees = set([(decree[0:4], decree[5:8], decree[8:]) for decree in decree_number_regex.findall(text)])
    new_decrees = decrees.difference(mentioned_decrees)
    mentioned_decrees = mentioned_decrees.union(new_decrees)
    decrees_dict[decree_date] = new_decrees

series = sorted(set((y, s) for (y, s, n) in mentioned_decrees if y >= min_year))
print(';'.join([''] + sorted(decrees_dict.keys())))

for (y, s) in series:
    def belongs_cur_series(x):
        cur_y, cur_s, _ = x
        return y == cur_y and s == cur_s

    def cur_series_count_str(x):
        result = len(list(filter(belongs_cur_series, x)))
        return str(result) if result else ''

    print(';'.join([f'{y}X{s}'] + [cur_series_count_str(v) for v in decrees_dict.values()]))