from pprint import PrettyPrinter
from urllib.parse import quote

import re
import requests
import xmltodict

def search_questions(query='', limit=1):
    query = quote(query)
    url = 'https://db.chgk.info/xml/search/questions/{}/Q/limit{}'.format(query, limit)
    response = xmltodict.parse(requests.get(url).text)

    if 'question' not in response['search']:
        return

    entities = response['search']['question']
    if not isinstance(entities, list):
        entities = [entities]

    for entity in entities:
        for field in ['Comments', 'Question']:
            data = entity[field]
            if data is None:
                continue

            regex = r'\s*\(pic: [^)]+\)\s*'
            matches = re.findall(regex, data)
            if not matches:
                continue

            images = []
            image_base_url = 'https://db.chgk.info/images/db/'
            regex_group = r'\s*\(pic: ([^)]+)\)\s*'
            for match in matches:
                url = re.match(regex_group, match)[1]
                if not url.startswith('http'):
                    url = image_base_url + url
                images.append(url)

            entity[field + 'Images'] = images
            entity[field] = re.sub(regex, '', data).strip()

    return entities

entities = search_questions('pic', 1)
PrettyPrinter(width=100).pprint(entities)