A utility to fetch all countries' captial.

captial-crawler.py

#!/usr/bin/env python
# encoding: utf-8

import urllib2, re, json, random

url = 'http://www.tripmondo.com/magazine/facts-and-statistics/list-of-capitals-and-countries/'
html = urllib2.urlopen(url, timeout=5).read()
reg = re.compile('<td><a href=.*?>(.*?)</a></td>\s+<td><a href=.*?>(.*?)</a></td>\s+<td class="right">(.*?)</td>\s+<td class="right">(.*?)</td>')

with open('city.json', 'w') as f :
    f.write("[\n")
    matches = reg.findall(html)
    for country, captial, latitude, longitude in matches :
        wiki = 'http://en.wikipedia.org/wiki/' + captial.replace(' ', '_')
        lat = float(latitude.replace(',', '.'))
        log = float(longitude.replace(',', '.'))
        f.write("\t[\"%s\", %.2f, %.2f, %.4f, \"%s\"],\n" % (captial, lat, log, random.random(), wiki))
        print country, captial, latitude, longitude
    f.write("]\n")