src/resources/create_city_urls_json.py

import json
from pathlib import Path

p = Path(__file__).with_name('large_cities.txt')

urls = []
# with p.open('r') as f:
#     for line in f:
#         l = line.strip()
#         a = l.split()
#         j = {"url_type": "city_url", "url":a[2]}
#         urls.append(j)
def umlaut_to_ascii(text):
    return (text.replace("ä", "ae")
            .replace("ö", "oe")
            .replace("ü", "ue")
            .replace("Ä", "Ae")
            .replace("Ö", "Oe")
            .replace("Ü", "Ue")
            .replace("ß", "ss"))

with open('large_cities.txt', 'r', encoding='utf-8') as f:
    for line in f:
        l = line.strip()
        l = umlaut_to_ascii(l)  # Funktion zur Umwandlung von Umlauten
        print(l)

        a = l.split()
        j = {
            "url_type": "city",
            "url": f"https://www.{a[1]}.de",
            "meta": {
                "website_host": f"{a[1]}",
                "location": f"{a[1]}"
            }
        }
        urls.append(j)


urls.pop(0)
with open('large_city_urls.json', 'w', encoding='utf-8') as f:
    json.dump(urls, f, ensure_ascii=False, indent=4)