parse-ndf

#!/usr/bin/env python

from datetime import date
import json
import sys

from split.parse import unpack, cnv_full_date, loop, fare_file
from split.utils import ticket_types

lookups = (
    ('from', 4), ('to', 4), ('route', 5), ('railcard', 3), ('ticket', 3), ('type', 1), ('end_date', 8, cnv_full_date), ('start_date', 8, cnv_full_date), ('quote_date', 8, cnv_full_date), ('suppress', 1), ('adult', 8), ('child', 8), ('restriction', 2), ('composite', 1), ('cross-london', 1), ('private-settlement', 1)
)

def parse_file(file, type):
    for row in loop(file):
        row = unpack(row, *lookups)
        assert row['type'] == type
        if row['railcard'].strip(): continue
        if row['end_date'] < date.today(): continue
        if row['start_date'] > date.today(): continue
        if row['ticket'] not in ticket_types: continue
        if row['composite'] == 'N': continue
        del row['railcard'], row['end_date'], row['quote_date'], row['start_date'], row['composite'], row['child'], row['cross-london'], row['private-settlement'], row['type']
        key = '%(from)s-%(to)s-%(route)s-%(ticket)s' % row
        yield key, row

# ---

data = {}
for key, row in parse_file(fare_file('NFO'), 'O'):
    assert row['suppress'] == 'N'
    del row['suppress']
    data[key] = row

print("%d rows from NFO file" % len(data), file=sys.stderr)

json.dump(data, open('data/ndf.json', 'w'), separators=(',', ': '), indent=2, sort_keys=True)