Automatically exported from code.google.com/p/planningalerts
Você não pode selecionar mais de 25 tópicos Os tópicos devem começar com uma letra ou um número, podem incluir traços ('-') e podem ter até 35 caracteres.

KingstonUponThames.py 2.7 KiB

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667686970
  1. import urllib2
  2. import urllib
  3. import urlparse
  4. import datetime, time
  5. import cgi
  6. from BeautifulSoup import BeautifulSoup
  7. from PlanningUtils import PlanningApplication, \
  8. PlanningAuthorityResults, \
  9. getPostcodeFromText
  10. date_format = "%d/%b/%Y"
  11. class KingstonParser:
  12. comments_email_address = "dc@rbk.kingston.gov.uk"
  13. def __init__(self, *args):
  14. self.authority_name = "Royal Borough of Kingston upon Thames"
  15. self.authority_short_name = "Kingston upon Thames"
  16. self.base_url = "http://maps.kingston.gov.uk/isis_main/planning/planning_summary.aspx?strWeekListType=SRCH&strRecTo=%(date)s&strRecFrom=%(date)s&strWard=ALL&strAppTyp=ALL&strWardTxt=All%%20Wards&strAppTypTxt=All%%20Application%%20Types&strStreets=ALL&strStreetsTxt=All%%20Streets&strLimit=500"
  17. self._results = PlanningAuthorityResults(self.authority_name, self.authority_short_name)
  18. def getResultsByDayMonthYear(self, day, month, year):
  19. search_day = datetime.date(year, month, day)
  20. # Now get the search page
  21. response = urllib2.urlopen(self.base_url %{"date": search_day.strftime(date_format)})
  22. soup = BeautifulSoup(response.read())
  23. # Each app is stored in a table on it's own.
  24. # These tables don't have any nice distinguishing features,
  25. # but they do all contain a NavigableString "Application",
  26. # and nothing else in the page does.
  27. nav_strings = soup.findAll(text="Application")
  28. for nav_string in nav_strings:
  29. results_table = nav_string.findPrevious("table")
  30. application = PlanningApplication()
  31. application.date_received = search_day
  32. application.council_reference = results_table.a.string.strip()
  33. application.info_url = urlparse.urljoin(self.base_url, results_table.a['href'])
  34. application.address = results_table.findAll("td")[7].a.string.strip()
  35. application.postcode = getPostcodeFromText(application.address)
  36. application.description = results_table.findAll("td")[-1].contents[0].strip()
  37. # A few applications have comment urls, but most don't.
  38. # When they do, they have a case officer - I don't think we can
  39. # work out the other urls - even if they exist.
  40. # Best to use the email address.
  41. application.comment_url = self.comments_email_address
  42. self._results.addApplication(application)
  43. return self._results
  44. def getResults(self, day, month, year):
  45. return self.getResultsByDayMonthYear(int(day), int(month), int(year)).displayXML()
  46. if __name__ == '__main__':
  47. parser = KingstonParser()
  48. print parser.getResults(2,8,2008)