python - logs into email and prints your oyster card costs per month by looking for the 'Oyster card topped up' messages in your TFL_MAILBOX email folder

tfl_email_costs.py

import imaplib
import email
from pprint import pprint
import re
import collections
from email.Iterators import typed_subpart_iterator

""" Assumes you have TFL email you each time your oyster auto top up occurs """


USERNAME = ""  # Gmail user
PASSWORD = ""  # Gmail pass
TFL_MAILBOX = "good mailshots/trains/tfl"  # Gmail label where your TFL emails go to

def get_charset(message, default="ascii"):
    """Get the message charset"""

    if message.get_content_charset():
        return message.get_content_charset()

    if message.get_charset():
        return message.get_charset()

    return default

def get_body(message):
    """Get the body of the email message"""

    if message.is_multipart():
        #get the plain text version only
        text_parts = [part
                      for part in typed_subpart_iterator(message,
                                                         'text',
                                                         'plain')]
        body = []
        for part in text_parts:
            charset = get_charset(part, get_charset(message))
            body.append(unicode(part.get_payload(decode=True),
                                charset,
                                "replace"))

        return u"\n".join(body).strip()

    else: # if it is not multipart, the payload will be a string
          # representing the message body
        body = unicode(message.get_payload(decode=True),
                       get_charset(message),
                       "replace")
        return body.strip()



money_pattern = re.compile("was topped up with .?(\d+)")
money_pattern_html = re.compile("was topped up with £(\d+)")
def extract_money(payload):
    return int(money_pattern.search(payload).group(1))


conn = imaplib.IMAP4_SSL("imap.gmail.com", 993)
conn.login(USERNAME, PASSWORD)
conn.select()
try:
    conn.select(TFL_MAILBOX)
    typ, msg_ids = conn.search(None, '(SUBJECT "Your Oyster card has been topped up")')

    result = []

    for num in msg_ids[0].split():
        typ, msg_data = conn.fetch(num, '(RFC822)')
        for response_part in msg_data:
            if isinstance(response_part, tuple):
                msg = email.message_from_string(response_part[1])
                body = get_body(msg)

                try:
                    result.append( (msg.get('date'), extract_money(body)) )
                except Exception, e:
                    print e
                    print body
                    print 'bad one on day: ' + msg.get('date')

    r_map = collections.OrderedDict()

    for r in result:
        index = r[0].split()[3] + r[0].split()[2]
        r_map.setdefault(index, 0)
        r_map[index] += r[1]

    for r,v in r_map.items():
        print r + ':' + str(v)


finally:
    try:
        conn.close()
    except:
        pass
    conn.logout()



# list_response_pattern = re.compile(r'\((?P<flags>.*?)\) "(?P<delimiter>.*)" (?P<name>.*)')
# def parse_list_response(line):
#     flags, delimiter, mailbox_name = list_response_pattern.match(line).groups()
#     mailbox_name = mailbox_name.strip('"')
#     return (flags, delimiter, mailbox_name)


# def view_mailboxes(conn):
#     typ, data = conn.list()
#     for line in data:
#         flags, delimiter, mailbox_name = parse_list_response(line)
#         print conn.status(mailbox_name, '(MESSAGES RECENT UIDNEXT UIDVALIDITY UNSEEN)')