metasov · October 29, 2012 02:56
diff --git a/vibory-vibory.py b/vibory-vibory.py
 #!/usr/bin/env python
 # -*- coding: utf-8

 import curl
 from datetime import datetime

 connection = curl.Curl()
 connection.set_timeout(20)

 def get_document(url):
 	import time
 	from lxml.html import document_fromstring

 	html = ""
 	while not html:
 		html = ""
 		try:
 			data = connection.get(url)
 			html = data.decode('cp1251')
 			if not html:
 				raise Exception("Error: Blank Output")
 			if html.find(u"Інформація щодо цієї сторінки відсутня.")!=-1:
 				html = ""
 				raise Exception("Error: Інформація щодо цієї сторінки відсутня")
 		except Exception, e:
 			print e
 			time.sleep(5)

 	doc = document_fromstring(html)
 	return doc

 def get_results_by_ovk(ovk_num):
 	results = {}
 	
 	doc = get_document('http://cvk.gov.ua/vnd2012/wp304pt001f01=900pf7331=%d.html'%ovk_num)
 	rows = doc.xpath('//div[@id="content"]/table[@class="t2"][2]//tr')[1:]

 	for row in rows:
 		name, null, percent, count = map( lambda x:x.text_content().strip(), row.getchildren() )
 		percent, count = float(percent)/100, int(count.replace(' ', ''))
 		results.setdefault(name, 0)
 		results[name] = percent

 	return results


 doc = get_document('http://cvk.gov.ua/vnd2012/wp300pt001f01=900.html')
 overall_percent = doc.xpath('//*[@id="content"]/table[4]//tr/td[2]')[0].text_content().replace(u' ', u'')

 ovks = {} 

 doc = get_document('http://cvk.gov.ua/vnd2012/wp095_2pt001f01=900pt049f01=4.html')
 rows = doc.xpath('//div[@id="content"]/table[@class="t2"][2]//tr')[1:-1]
 for row in rows:
 	name, _all_count, real_count = map( lambda x:x.text_content().strip(), row.getchildren() )
 	num = int(name.replace(u'ОВО №', u''))
 	count = int(real_count.replace(u' ', u''))
 	ovks[num] = count

 ovks_results = {}

 for ovk in ovks.keys():
 	ovks_results[ovk] = get_results_by_ovk(ovk)

 count_by_parties = {}
 all_count = 0

 for ovk in ovks.keys():
 	results = ovks_results[ovk]
 	for party in results.keys():
 		count_by_parties.setdefault(party, 0)
 		count = results[party]*ovks[ovk]
 		count_by_parties[party] += count
 		all_count += count

 results = []
 for party in count_by_parties.keys():
 	results.append( (party, count_by_parties[party]/all_count*100) )

 results.sort(key = lambda x:x[1], reverse=True)

 print u"<html><head><title>Выборы-Выборы...</title><meta http-equiv='Content-Type' content='text/html; charset=utf-8'/></head>"
 print u"<body>"
 print u"Обновлено: %s<br />"%datetime.now()
 print u"Обработано протоколов: %s<br />"%overall_percent
 print u"<table>"
 for result in results:
 	print u"<tr><td>%s</td> <td>%f</td></tr>"%result
 print u"</table></body></html>"
	#!/usr/bin/env python
	# -*- coding: utf-8

	import curl
	from datetime import datetime

	connection = curl.Curl()
	connection.set_timeout(20)

	def get_document(url):
	import time
	from lxml.html import document_fromstring

	html = ""
	while not html:
	html = ""
	try:
	data = connection.get(url)
	html = data.decode('cp1251')
	if not html:
	raise Exception("Error: Blank Output")
	if html.find(u"Інформація щодо цієї сторінки відсутня.")!=-1:
	html = ""
	raise Exception("Error: Інформація щодо цієї сторінки відсутня")
	except Exception, e:
	print e
	time.sleep(5)

	doc = document_fromstring(html)
	return doc

	def get_results_by_ovk(ovk_num):
	results = {}

	doc = get_document('http://cvk.gov.ua/vnd2012/wp304pt001f01=900pf7331=%d.html'%ovk_num)
	rows = doc.xpath('//div[@id="content"]/table[@class="t2"][2]//tr')[1:]

	for row in rows:
	name, null, percent, count = map( lambda x:x.text_content().strip(), row.getchildren() )
	percent, count = float(percent)/100, int(count.replace(' ', ''))
	results.setdefault(name, 0)
	results[name] = percent

	return results


	doc = get_document('http://cvk.gov.ua/vnd2012/wp300pt001f01=900.html')
	overall_percent = doc.xpath('//*[@id="content"]/table[4]//tr/td[2]')[0].text_content().replace(u' ', u'')

	ovks = {}

	doc = get_document('http://cvk.gov.ua/vnd2012/wp095_2pt001f01=900pt049f01=4.html')
	rows = doc.xpath('//div[@id="content"]/table[@class="t2"][2]//tr')[1:-1]
	for row in rows:
	name, _all_count, real_count = map( lambda x:x.text_content().strip(), row.getchildren() )
	num = int(name.replace(u'ОВО №', u''))
	count = int(real_count.replace(u' ', u''))
	ovks[num] = count

	ovks_results = {}

	for ovk in ovks.keys():
	ovks_results[ovk] = get_results_by_ovk(ovk)

	count_by_parties = {}
	all_count = 0

	for ovk in ovks.keys():
	results = ovks_results[ovk]
	for party in results.keys():
	count_by_parties.setdefault(party, 0)
	count = results[party]*ovks[ovk]
	count_by_parties[party] += count
	all_count += count

	results = []
	for party in count_by_parties.keys():
	results.append( (party, count_by_parties[party]/all_count*100) )

	results.sort(key = lambda x:x[1], reverse=True)

	print u"<html><head><title>Выборы-Выборы...</title><meta http-equiv='Content-Type' content='text/html; charset=utf-8'/></head>"
	print u"<body>"
	print u"Обновлено: %s<br />"%datetime.now()
	print u"Обработано протоколов: %s<br />"%overall_percent
	print u"<table>"
	for result in results:
	print u"<tr><td>%s</td> <td>%f</td></tr>"%result
	print u"</table></body></html>"