Print a frequency distribution of countries for a list of IP addresses using a local GeoIP database

country_count.py

#!/usr/bin/env python
import argparse
import json
import re
import sys

import GeoIP

from collections import Counter


GEOCODER = GeoIP.open("/usr/local/share/GeoIP/GeoIP.dat", GeoIP.GEOIP_STANDARD)

def perform(args):
    country_map = Counter()
    for line in args.input_file:
        if re.search(r'^\s*[0-9]+\.[0-9]+\.[0-9]+\.[0-9]+\s*$', line):
            ipaddr = line.strip()
            country = GEOCODER.country_code_by_addr(ipaddr)
            if args.only:
                if country in args.only:
                    country_map[country] += 1
                else:
                    country_map['*'] += 1
            else:
                country_map[country] += 1
    for key, value in country_map.iteritems():
        args.output_file.write('%s\t%s\n' % (key, value))

def main():
    ap = argparse.ArgumentParser()
    ap.add_argument('input_file', nargs='?', type=argparse.FileType('r'), default=sys.stdin)
    ap.add_argument('output_file', nargs='?', type=argparse.FileType('w'), default=sys.stdout)
    ap.add_argument('--only', type=str, nargs='+', default=[])
    args = ap.parse_args()
    perform(args)

if __name__ == '__main__':
    main()