Plot measles epidemic in Serbia

measles.py

# -*- coding: utf-8 -*-
import requests
import bs4
import re
import datetime
from matplotlib import pyplot

replacements = str.maketrans({
 'а':'a',
 'е':'e',
 'ј':'j',
 'к':"k",
 'м':"m",
 'о':"o",
 'т':"t",
 'А':'A',
 'Е':'E',
 'Ј':'J',
 'К':"K",
 'М':"M",
 'О':"O",
 'Т':"T"})

url = "http://www.batut.org.rs/index.php?content=1629"
proxies={"http":"socks5://localhost:8081"}

def get_text(url, proxies):
    r = requests.get(url, proxies=proxies)
    data = r.content.decode("utf8")
    soup = bs4.BeautifulSoup(data, "html.parser")
    text = soup.find_all(attrs={"class":"news_content_full"})[0].get_text()
    text = text.replace("\r\n", "\n").replace("\n\n","")
    text = text.translate(replacements)
    return text


def get_first_paragraphs(text):
    start = "Od početka oktobra 2017"
    num_re = re.compile("ukupno (\\d+) sluča")
    return [x for x in text.splitlines() if start in x and len(num_re.findall(x))==1]

date_re = re.compile("\\d{1,2}\\.\\d{1,2}\\.\\s{,3}\\d{2,4}\\.")
def extract_date(s):
    global date_re
    d = date_re.findall(s)[0].replace(" ", "")
    return datetime.datetime.strptime(d, "%d.%m.%Y.")

def extract(paragraphs):
    num_re = re.compile("ukupno (\\d+) sluča")    
    return [(extract_date(x), int(num_re.findall(x)[0])) for x in paragraphs]

def get_data(url, proxies):    
    data = get_first_paragraphs(get_text(url, proxies))
    result = extract(data);
    return result

if __name__ == "__main__":
    d = get_data(url, proxies)
    x,y = zip(*d)
    
    fig = pyplot.figure()    
    ax = fig.add_subplot(111)
    ax.set_title("Epidemija morbila u Srbiji", fontsize=18, fontweight="bold")
    ax.plot(x,y)
    
    ax.text(1.0, 0, "Total: {0}".format(y[0]), verticalalignment='bottom', 
            horizontalalignment='right', transform=ax.transAxes, fontsize=14)
    pyplot.show();