age_scraper.py

#encoding=utf-8
import requests
import re


data = []
regions = []

r = requests.get('http://www.lv.ukrstat.gov.ua/dem/SWF/dani/age/age1.php')
r.encoding = 'cp1251'
lines = r.text.splitlines()
for line in lines:
  region = re.search('<option value="(.*?)"(?: selected )?>(.*?)</option>', line)
  if region:
    regions.append(region.groups())


for id_,name in regions:
  print name
  r = requests.get('http://www.lv.ukrstat.gov.ua/dem/SWF/dani/age/%s.svg' % id_)
  r.encoding = 'cp1251'
  lines = r.text.splitlines()
  for line in lines:
    years_match = re.search('var currYear = \[(.*?)\];', line)
    if years_match:
      years = map(lambda x: x.replace("'", ''), years_match.groups()[0].split(','))
  for line in lines:
    age_match = re.search('stat([a-z]+)(\d\d?\d?)=\[(.*?)\];', line)
    if age_match:
      sex, age, count_by_year = age_match.groups()
      counts_by_year = count_by_year.replace('"', '').split(',')[1:]
      for i, count in enumerate(counts_by_year):
        data.append((id_, name.strip(), sex, age, years[i], count))


with open('age_stats.csv', 'w') as f:
  f.write((','.join([u'КОАТУУ регіону',u'регіон',u'стать',u'вік',u'рік',u'кількість']) + '\n').encode('utf-8'))
  for line in data:
    f.write((','.join(line) + '\n').encode('utf-8'))

И че ? работает скрипт ?