filter.py

#!/usr/bin/env python

enddays = [0, 31,28,31,30,31,30,31,31,30,31,30,31]

def output_transactions(saldo, days):
    this_year = 2010 #datetime.datetime.now().year
    dtstart = datetime.date(this_year, days.keys()[1].month, 1)
    dtend = datetime.date(this_year, dtstart.month, enddays[dtstart.month])
    for day in daterange(dtstart, dtend):
        if day in days.keys():
            for trans in days[day]:
                action = trans['action'].split('\t')[0]
                value = trans['value']
                print ("%s\t%s\t%.02f" % (day, action, value)).replace('.', ',')

def daterange(start_date, end_date):
    for n in range((end_date - start_date).days+1):
        yield start_date + datetime.timedelta(n)

def money(x):
    x = x.split()
    x[0] = x[0].replace('.', '')
    x[0] = x[0].replace(',', '.')
    return +float(x[0]) if x[1] == 'C' else -float(x[0])

def todate(x):
    if not type(x) is type(''):
        return x
    day = int(x.split('/')[0])
    month = int(x.split('/')[1])
    year = 2010
    return datetime.date(year, month, day)

import sys, re, datetime, collections

lines = [x.strip() for x in sys.stdin.readlines()]

blck_start = lines.index('Movimentacao em contas')
blck_end = lines.index('', blck_start)

curday = ''
days = collections.defaultdict(list)
saldo = 0

for l in lines[blck_start:blck_end]:
    m = re.search('^(\\d{2}/\\d{2})', l)
    if m:
        curday = m.groups()[0]
        action = l[6:41]
        value = l[42:]
    elif not l.startswith('Saldo Total'):
        action = l[0:35]
        value = l[36:]
    else:
        continue

    # once per history
    if action.startswith('Saldo anterior'):
        saldo = money(action.split('\t')[-1])
        continue

    if not curday:
        continue

    value = money(value)
    curday = todate(curday)

    days[curday] += [{"action":action, "value":value}]

output_transactions(saldo, days)