Created
November 15, 2021 10:28
-
-
Save scharf/10f5294d8443e7871d63f97a9a6653a4 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import re | |
import xml.etree.ElementTree as ET | |
import sys | |
# from pprint import pprint | |
from collections import ChainMap | |
def xml_to_dict_recursive(root): | |
if len(root.getchildren()) == 0: | |
return {root.tag: root.text} | |
else: | |
children = list(map(xml_to_dict_recursive, root.getchildren())) | |
x = set([c for d in children for c in d.keys()]) | |
if len(set([c for d in children for c in d.keys()])) == len(children): | |
children = dict(ChainMap(*children)) | |
else: | |
pass | |
return {root.tag: children} | |
def main(): | |
infile = sys.argv[1] | |
tree = ET.parse(infile) | |
data = xml_to_dict_recursive(tree.getroot()) | |
# pprint(data, indent=4) | |
contacts = [c for c in data['phonebooks']['phonebook']] | |
contacts = [c['contact'] for c in contacts] | |
# pprint(data['phonebooks'][0]['phonebook']) | |
# pprint(contacts, indent=4) | |
with open(re.sub('\.xml$', '.txt', infile), 'w') as file: | |
print( | |
'"Name","Vorname","Rufnummer Privat","Rufnummer Arbeit","Rufnummer Mobil","Rufnummer Mobil 2","Strasse, Nr.","PLZ","Ort","Geburtstag"', | |
file=file) | |
for c in contacts: | |
namen = c['person']['realName'].split(', ') | |
if len(namen) > 2: | |
raise RuntimeError() | |
if len(namen) == 1: | |
vorname = '' | |
nachname = namen[0] | |
else: | |
nachname, vorname = namen | |
numbers = c['telephony'] | |
priv1 = '' | |
priv2 = '' | |
mobil1 = '' | |
mobil2 = '' | |
if isinstance(numbers, dict): | |
number = numbers['number'] | |
if number.startswith('01'): | |
mobil1 = number | |
else: | |
priv1 = number | |
else: | |
for number in [n['number'] for n in numbers]: | |
if number.startswith('01'): | |
if not mobil1: | |
mobil1 = number | |
else: | |
mobil2 = number | |
else: | |
if not priv1: | |
priv1 = number | |
else: | |
priv2 = number | |
def fix_tel(n): | |
n = re.sub(r'\s+', '', n) | |
if not re.match(r'^[\d\s]+$', n): | |
n = '' | |
return n | |
priv1, priv2, mobil1, mobil2 = [fix_tel(n) for n in [priv1, priv2, mobil1, mobil2]] | |
if not ''.join([priv1, priv2, mobil1, mobil2]) or not ''.join([nachname, vorname]): | |
print(c) | |
else: | |
print(f'"{nachname}","{vorname}","{priv1}","{priv2}","{mobil1}","{mobil2}","","","",""', file=file) | |
if __name__ == '__main__': | |
main() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment