loa_federal_rdf_to_csv.py

# Converte o arquivo RDF em http://dados.gov.br/dataset/orcamento-federal/resource/edf2263a-f8d9-48d9-b25d-6863d16ce0b0?inner_span=True
import re

loa_file = open('./loa.items')
numero_items = 375960
item_re = re.compile(r'<[^>]+/([^>]+)>.*/([^>]+)>')
item_value_re = re.compile(r'<[^>]+/([^>]+)>.* "(.*)"\^\^.*')

s = ''
for i in range(numero_items):
    sp = ''
    for j in range(9):
        line = loa_file.readline().rstrip()[:-2]
        sep = ', ' if j != 0 and j != 9 else ''
        m = item_re.match(line)
        if j == 0:
            sp += m.group(1) + ', '
        sp += '{}{}'.format(sep, m.group(2))
    for j in range(6):
        line = loa_file.readline().rstrip()[:-2]
        sep = ', ' if j != 6 else '\n'
        m = item_value_re.match(line)
        sp += '{}{}'.format(sep, m.group(2))
    loa_file.readline()
    s += sp + '\n'

output = open('loa.csv', 'w')
output.write(s)