First implementation of parser for Yandex`s result pages

reps.py

from lxml import html

class YandexRespParser(object):

    __RESULTS_ON_PAGE = 10 #Its possible to evaluate this number from first result page
    
    def __create_rep(self, result_element):
        rep = {}
        
        rep["index"] = int(result_element.xpath('.//b[@class="b-serp-item__number"]')[0].text_content())
        rep["href"] = str(result_element.xpath('.//a[@class="b-serp-item__title-link"]')[0].get("href"))
        rep["domain"] = str(result_element.xpath('.//a[@class="b-serp-url__link"]')[0].text_content())
        rep["title"] = str(result_element.xpath('.//a[@class="b-serp-item__title-link"]')[0].text_content())
        rep["text"] = str(result_element.xpath('.//div[@class="b-serp-item__text"]')[0].text_content())
        rep["copy"] = str(result_element.xpath('.//a[@class="b-serp-item__links-link"]')[0].get("href"))
        
        return rep
        
    def __get_reps_from_page(self, page_number, results_from_page=10):
        page = html.parse("http://ya.ru/yandsearch?text={0}&lr={1}&p={2}".format(self.query, self.lr, page_number))
        result_elements = page.xpath('//li[@class="b-serp-item i-bem"]')
        
        reps_count = min(results_from_page, len(result_elements))
        
        return [self.__create_rep(result_elements[i]) for i in range(reps_count)] 
                        
    def get_reps (self, count):
        number_of_pages = count / self.__RESULTS_ON_PAGE
        last_results = count % self.__RESULTS_ON_PAGE
        
        first_search_results = sum(map(self.__get_reps_from_page, range(number_of_pages)), [])
        
        last_search_results = self.__get_reps_from_page(number_of_pages, last_results)
        
        return first_search_results + last_search_results
   
    def __init__(self, query, lr):
        self.query = query
        self.lr = lr