fix_term_start_dates.py

import csv
import datetime
import itertools
from utils import load_data, save_data

terms = []
for row in csv.DictReader(open("sessions.tsv"), delimiter='\t'):
    terms.append(row)

ordinal = lambda n: "%d%s" % (n,"tsnrhtdd"[(n/10%10!=1)*(n%10<4)*n%10::4])

replacements = {}
for k, v in itertools.groupby(terms, lambda x: x['congress']):
    sessions = list(v)
    special = [x for x in sessions if x['session'] == 'A']
    if len(special) == 0:
        continue
    special = special[0]
    first = [x for x in sessions if x['session'] == '1'][0]

    special_start = datetime.datetime.strptime(special['start'], '%Y-%m-%d').date()
    first_start = datetime.datetime.strptime(first['start'], '%Y-%m-%d').date()

    if (special_start < first_start):
        replacements[str(first_start)] = str(special_start)
        print("{:>4} Congress {} => {}".format(ordinal(int(k)), first_start, special_start))

y = load_data("legislators-historical.yaml")

for m in y:
    for t in m['terms']:
        if t['start'] in replacements:
            t['start'] = replacements[t['start']]

save_data(y, "legislators-historical.yaml")