main.py

import re
import json
from datetime import datetime

import requests
from bs4 import BeautifulSoup

base_url = "https://index.minfin.com.ua/ua/russian-invading/casualties"


def get_url():
    response = requests.get(base_url)
    soup = BeautifulSoup(response.text, 'html.parser')
    content = soup.select('div[class=ajaxmonth] h4[class=normal] a')
    urls = ["/"]
    prefix = "/month.php?month="
    for tag_a in content:
        urls.append(prefix + re.search(r"\d{4}-\d{2}", tag_a["href"]).group())
    return urls


def spider(urls):
    data = []
    for url in urls:
        response = requests.get(base_url + url)
        soup = BeautifulSoup(response.text, 'html.parser')
        content = soup.select('ul[class=see-also] li[class=gold]')
        for element in content:
            result = {}
            date = element.find('span', attrs={"class": "black"}).text
            try:
                date = datetime.strptime(date, "%d.%m.%Y").isoformat()
            except ValueError:
                print(f"error for {date}")
                continue

            result.update({"date": date})
            losses = element.find('div').find('div').find('ul')
            for l in losses:
                title, quantity, *rest = l.text.split('—')
                title = title.strip()
                quantity = re.search(r"\d+", quantity).group()
                result.update({title: quantity})
            data.append(result)
    return data


if __name__ == '__main__':
    urls_for_parser = get_url()
    r = spider(urls_for_parser)
    with open('moskali.json', 'w', encoding='utf-8') as fd:
        json.dump(r, fd, ensure_ascii=False)