from collections import defaultdict

all_values = []
age_value_map = defaultdict(list)

with open('2015_2017_MaleData.dat') as f:

    for line in f.readlines():

        # note: columns in data sheet definition
        # are 1-based, hence the -1 in indices

        # data sheet defines AGE_A(13-14)
        # as the age of the respondend

        age = int(line[12:14])

        # data sheet defines MON12PRTS(157)
        # as "Number of female sex partners in last 12 months"

        number_str = line[156]

        if number_str == ' ':
            # ignore N/A data
            continue

        number = int(number_str)
        all_values.append(number)
        age_value_map[age].append(number)


def calc_median(values):
    middle = len(values) // 2
    sorted_values = sorted(values)
    return sorted_values[middle]


def age_group_values(age_value_map, youngest, oldest):
    group_values = []
    for age in range(youngest, oldest + 1):
        if age in age_value_map:
            group_values.extend(age_value_map[age])
    return group_values


print("Overall median: {}".format(calc_median(all_values)))

group_18_24 = age_group_values(age_value_map, 18, 24)
print("Median of group 18-25: {}".format(calc_median(group_18_24)))
group_25_34 = age_group_values(age_value_map, 25, 34)
print("Median of group 25-34: {}".format(calc_median(group_25_34)))
group_35_44 = age_group_values(age_value_map, 35, 44)
print("Median of group 35-44: {}".format(calc_median(group_35_44)))
group_45_54 = age_group_values(age_value_map, 45, 54)
print("Median of group 35-44: {}".format(calc_median(group_45_54)))