Kontonummer / BLZ to IBAN / BIC converter

sepa.py

# coding: utf-8
# author: Lorenz H-S (github.com/lorenzhs)
# license: 3-clause BSD

import re
import urllib2
from datetime import date
import codecs
from collections import defaultdict
import sys

def knrBlzToIBAN(knr, blz):
    bban = int(str(blz) + str(knr).zfill(10) + "131400") # 1314 = DE
    remainder = 98 - (bban % 97)
    checksum = str(remainder).zfill(2)
    iban = "DE" + checksum + str(blz) + str(knr).zfill(10) 
    return iban

def downloadLatestBlzBICTable(outfilename):
    BASE_URL = "http://www.bundesbank.de"
    overview_url = "{base}/Redaktion/DE/Standardartikel/Kerngeschaeftsfelder/Unbarer_Zahlungsverkehr/bankleitzahlen_download.html".format(base=BASE_URL)
    datafile_regexp = re.compile('<a href="(.*/blz_([0-9]{4})_([0-9]{2})_([0-9]{2})_txt\.txt.*)"')
    print 'Dowloading list of available tables...'
    overview = urllib2.urlopen(overview_url).read()
    datafiles = datafile_regexp.findall(overview)

    max_date = date(1970,1,1)
    latest_url = None
    for (url, year, month, day) in datafiles:
        curr_date = date(int(year), int(month), int(day))
        if curr_date > max_date and curr_date <= date.today():
            latest_url = url
    if latest_url is None:
        print "Error: No presently valid data files found"
    else:
        latest_url = BASE_URL + latest_url

    print "Downloading table from", latest_url,
    sys.stdout.flush()
    data = urllib2.urlopen(latest_url).read()
    print "done."
    with open(outfilename, 'w') as outfile:
        outfile.write(data)

def parseBlzBICTable(infilename):
    with codecs.open(infilename, 'r', 'iso-8859-1') as infile:
        data = infile.readlines()
    lines = [line.strip() for line in data]
    
    blzDir = defaultdict(list)
    for line in lines:
        blz = int(line[:8])
        hauptniederlassung = bool(int(line[8]))
        #if not hauptniederlassung:
        #    continue
        bank_name = line[9:66].strip()
        plz = int(line[67:72])
        location = line[72:106].strip()
        branch_name = line[107:134].strip()
        bic = line[139:150].strip()

        if bic != '':
            blzDir[blz].append((bic, hauptniederlassung, bank_name, plz, location, branch_name))
    return blzDir

def writeBlzBICTableToHumanReadableFormat(blzDir, outfilename):
    with codecs.open(outfilename, 'w', 'utf-8') as f:
        for (key, values) in blzDir.iteritems():
            f.write(unicode(key))
            if len(values) > 1:
                f.write('\n')
            for val in values:
                f.write(u'\t' + u'\t'.join([unicode(v) for v in val]) + '\n')

if __name__ == '__main__':
    downloadLatestBlzBICTable('blzbictable')
    print 'parsing table...'
    blzDir = parseBlzBICTable('blzbictable')
    print 'parsed', len(blzDir), 'banks\' details'
    writeBlzBICTableToHumanReadableFormat(blzDir, 'blz_to_bic')