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.
 
 
 
 
 
 

210 linhas
8.9 KiB

  1. import urllib2
  2. import urllib
  3. import urlparse
  4. import datetime, time
  5. import cookielib
  6. cookie_jar = cookielib.CookieJar()
  7. from BeautifulSoup import BeautifulSoup
  8. from PlanningUtils import PlanningApplication, \
  9. PlanningAuthorityResults, \
  10. getPostcodeFromText
  11. search_date_format = "%d-%m-%Y" # Format used for the accepted date when searching
  12. possible_date_formats = [search_date_format, "%d/%m/%Y"]
  13. class CookieAddingHTTPRedirectHandler(urllib2.HTTPRedirectHandler):
  14. """The standard python HttpRedirectHandler doesn't add a cookie to the new request after a 302. This handler does."""
  15. def redirect_request(self, req, fp, code, msg, headers, newurl):
  16. new_request = urllib2.HTTPRedirectHandler.redirect_request(self, req, fp, code, msg, headers, newurl)
  17. # We need to add a cookie from the cookie_jar
  18. cookie_jar.add_cookie_header(new_request)
  19. return new_request
  20. cookie_handling_opener = urllib2.build_opener(CookieAddingHTTPRedirectHandler())
  21. class OcellaParser:
  22. received_date_format = search_date_format
  23. def __init__(self,
  24. authority_name,
  25. authority_short_name,
  26. base_url,
  27. debug=False):
  28. self.authority_name = authority_name
  29. self.authority_short_name = authority_short_name
  30. self.base_url = base_url
  31. self.debug = debug
  32. self._results = PlanningAuthorityResults(self.authority_name, self.authority_short_name)
  33. # These will be used to store the column numbers of the appropriate items in the results table
  34. self.reference_col = None
  35. self.address_col = None
  36. self.description_col = None
  37. self.received_date_col = None
  38. self.accepted_date_col = None
  39. def getResultsByDayMonthYear(self, day, month, year):
  40. search_date = datetime.date(year, month, day)
  41. # First get the search page
  42. get_request = urllib2.Request(self.base_url)
  43. get_response = urllib2.urlopen(get_request)
  44. cookie_jar.extract_cookies(get_response, get_request)
  45. get_soup = BeautifulSoup(get_response.read())
  46. # We need to find where the post action goes
  47. action = get_soup.form['action']
  48. try:
  49. session_id = get_soup.find('input', {'name': 'p_session_id'})['value']
  50. except TypeError:
  51. # In the case of Middlesbrough, there is no session cookie,
  52. # but it seems we don't need it...
  53. session_id = None
  54. # # From Breckland
  55. # p_object_name=FRM_WEEKLY_LIST.DEFAULT.SUBMIT_TOP.01
  56. # p_instance=1
  57. # p_event_type=ON_CLICK
  58. # p_user_args=
  59. # p_session_id=53573
  60. # p_page_url=http%3A%2F%2Fwplan01.intranet.breckland.gov.uk%3A7778%2Fportal%2Fpage%3F_pageid%3D33%2C30988%26_dad%3Dportal%26_schema%3DPORTAL
  61. # FRM_WEEKLY_LIST.DEFAULT.START_DATE.01=02-06-2008
  62. # FRM_WEEKLY_LIST.DEFAULT.END_DATE.01=09-06-2008
  63. # FRM_WEEKLY_LIST.DEFAULT.PARISH.01=
  64. post_data = urllib.urlencode(
  65. [('p_object_name', 'FRM_WEEKLY_LIST.DEFAULT.SUBMIT_TOP.01'),
  66. ('p_instance', '1'),
  67. ('p_event_type', 'ON_CLICK'),
  68. ('p_user_args', ''),
  69. ('p_session_id', session_id),
  70. ('p_page_url', self.base_url),
  71. ('FRM_WEEKLY_LIST.DEFAULT.START_DATE.01', search_date.strftime(search_date_format)),
  72. ('FRM_WEEKLY_LIST.DEFAULT.END_DATE.01', search_date.strftime(search_date_format)),
  73. ('FRM_WEEKLY_LIST.DEFAULT.PARISH.01', ''),
  74. ]
  75. )
  76. post_request = urllib2.Request(action, post_data)
  77. cookie_jar.add_cookie_header(post_request)
  78. post_request.add_header('Referer', self.base_url)
  79. post_response = cookie_handling_opener.open(post_request)
  80. post_soup = BeautifulSoup(post_response.read())
  81. results_table = post_soup.find("table", summary="Printing Table Headers")
  82. trs = results_table.findAll("tr")
  83. # We'll use the headings in the first tr to find out what columns the address, description, etc are in.
  84. ths = trs[0].findAll("th")
  85. th_index = 0
  86. for th in ths:
  87. th_content = th.font.string.strip()
  88. if th_content == 'Reference' or th_content == 'Application Ref':
  89. self.reference_col = th_index
  90. elif th_content == 'Location':
  91. self.address_col = th_index
  92. elif th_content == 'Proposal':
  93. self.description_col = th_index
  94. elif th_content == 'Development Description':
  95. self.description_col = th_index
  96. elif th_content == 'Received Date' or th_content == 'Date Received':
  97. self.received_date_col = th_index
  98. elif th_content == 'Accepted Date':
  99. self.accepted_date_col = th_index
  100. th_index += 1
  101. # If there is a received date, we'll use that, otherwise, we'll have to settle for the accepted date.
  102. self.received_date_col = self.received_date_col or self.accepted_date_col
  103. # We want all the trs except the first one, which is just headers,
  104. # and the last, which is empty
  105. trs = trs[1:-1]
  106. for tr in trs:
  107. self._current_application = PlanningApplication()
  108. tds = tr.findAll("td")
  109. self._current_application.council_reference = (tds[self.reference_col].font.a or tds[self.reference_col].a.font).string.strip()
  110. date_string = tds[self.received_date_col]
  111. for possible_format in possible_date_formats:
  112. try:
  113. self._current_application.date_received = datetime.datetime(*(time.strptime(tds[self.received_date_col].font.string.strip(), possible_format)[0:6]))
  114. except ValueError:
  115. pass
  116. self._current_application.address = tds[self.address_col].font.string.strip()
  117. self._current_application.postcode = getPostcodeFromText(self._current_application.address)
  118. self._current_application.description = tds[self.description_col].font.string.strip()
  119. self._current_application.info_url = tds[self.reference_col].a['href']
  120. # This is what a comment url looks like
  121. # It seems to be no problem to remove the sessionid (which is in any case blank...)
  122. # I can't see a good way to avoid having to go to the info page to find the moduleid though.
  123. #http://wplan01.intranet.breckland.gov.uk:7778/pls/portal/PORTAL.wwa_app_module.link?p_arg_names=_moduleid&p_arg_values=8941787057&p_arg_names=_sessionid&p_arg_values=&p_arg_names=APPLICATION_REFERENCE&p_arg_values=3PL%2F2008%2F0877%2FF
  124. # For the moment, we'll just use the info url, as that seems to work.
  125. self._current_application.comment_url = self._current_application.info_url
  126. self._results.addApplication(self._current_application)
  127. return self._results
  128. def getResults(self, day, month, year):
  129. return self.getResultsByDayMonthYear(int(day), int(month), int(year)).displayXML()
  130. if __name__ == '__main__':
  131. # parser = OcellaParser("Arun", "Arun", "http://www.arun.gov.uk/iplanning/portal/page?_pageid=33,4139&_dad=portal&_schema=PORTAL")
  132. # parser = OcellaParser("Breckland Council", "Breckland", "http://wplan01.intranet.breckland.gov.uk:7778/portal/page?_pageid=33,30988&_dad=portal&_schema=PORTAL")
  133. # parser = OcellaParser("Ellesmere Port", "Ellesmere Port", "http://ocella.epnbc.gov.uk/portal/page?_pageid=33,38205&_dad=portal&_schema=PORTAL")
  134. # parser = OcellaParser("Fareham", "Fareham", "http://eocella.fareham.gov.uk/portal/page?_pageid=33,31754&_dad=portal&_schema=PORTAL")
  135. # parser = OcellaParser("Hillingdon", "Hillingdon", "http://w09.hillingdon.gov.uk/portal/page?_pageid=33,82093&_dad=portal&_schema=PORTAL")
  136. # parser = OcellaParser("North East Lincolnshire", "North East Lincolnshire", "http://planning.nelincs.gov.uk/portal/page?_pageid=33,68034&_dad=portal&_schema=PORTAL")
  137. # parser = OcellaParser("Middlesbrough", "Middlesbrough", "http://planserv.middlesbrough.gov.uk/portal/page?_pageid=33,4166&_dad=portal&_schema=PORTAL")
  138. # parser = OcellaParser("Uttlesford", "Uttlesford", "http://planning.uttlesford.gov.uk/portal/page?_pageid=33,35447&_dad=portal&_schema=PORTAL")
  139. # Bad status line? Try changing browser id string?
  140. # parser = OcellaParser("Bridgend", "Bridgend", "http://eplan.bridgend.gov.uk:7778/portal/page?_pageid=55,31779&_dad=portal&_schema=PORTAL")
  141. # Post never comes back
  142. # parser = OcellaParser("Havering", "Havering", "http://planning.havering.gov.uk/portal/page?_pageid=33,1026&_dad=portal&_schema=PORTAL")
  143. # Can't find the URL similar to the others, even though it is clearly Ocella
  144. # We get a 406 at the moment. Try browser id string?
  145. parser = OcellaParser("Great Yarmouth", "Great Yarmouth", "http://planning.great-yarmouth.gov.uk/portal/page/portal/plan/search")
  146. print parser.getResults(21,5,2008)
  147. #TODO
  148. # 1) Sort out proper comment url?
  149. # 2) Check for pagination
  150. # 3) Check no results case