Block ads and other annoyances by redirecting their host names to 127.0.0.1

adblock.py

#!/usr/bin/python

import requests

sources = [
  'http://pgl.yoyo.org/adservers/serverlist.php?hostformat=hosts&showintro=0&mimetype=plaintext',
  'http://hosts-file.net/.%5Cad_servers.txt',
  'https://adaway.org/hosts.txt',
  'http://winhelp2002.mvps.org/hosts.txt',
  'http://sysctl.org/cameleon/hosts',
  'http://someonewhocares.org/hosts/hosts'
]

headers = {
  'User-Agent': 'Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2228.0 Safari/537.36'
}

hosts = set()
hostnames = 0

for source in sources:
  print('# Fetching %s...' % source)
  request = requests.get(source, headers=headers)

  print('# Parsing %s...' % source)
  for line in (l.strip() for l in request.text.split('\n')):
    if not line.startswith('127.0.0.1'):
      continue
    if '#' in line:
      line = line.split('#', 2)[0].strip()
    ip_address, hostname = line.split()
    hostnames += 1
    hosts.add(hostname)

print('# %d hostnames, %d unique' % (hostnames, len(hosts)))
for hostname in sorted(hosts):
  print('127.0.0.1 %s' % hostname)