Plot the number of notices on the official site of Arch Linux

plot_arch.py

import urllib.request
from datetime import datetime
from collections import Counter
from threading import Thread
from bs4 import BeautifulSoup
import matplotlib.pyplot as plt


base_url_fmt = "https://www.archlinux.org/news/?page=%d"
max_page = 13


def fetch_page(page):
  print('Fetching page', page)
  with urllib.request.urlopen(base_url_fmt % page) as response:
    return response.read()


def main():
  years = []
  
  # Gather information
  for page in range(1, max_page+1):
    html = fetch_page(page)
    soup = BeautifulSoup(html, 'html5lib')
    years.extend(datetime.strptime(tr.td.contents[0], '%Y-%m-%d').date().year
                 for tr in soup.tbody.find_all('tr'))
  
  # How many years?
  max_year = max(years)
  min_year = min(years)
  
  # Each year, how many times?
  cnt = Counter(years)
  
  # Plot!
  x = list(range(min_year, max_year+1))
  y = [cnt[year] for year in x]
  plt.xlabel('Year')
  plt.ylabel('# Notices')
  plt.bar(x, y)
  plt.show()


if __name__ == '__main__':
  main()