Automatically exported from code.google.com/p/planningalerts
您最多选择25个主题 主题必须以字母或数字开头,可以包含连字符 (-),并且长度不得超过35个字符

AcolnetParser.py 17 KiB

17 年前
17 年前
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364
  1. #!/usr/local/bin/python
  2. import urllib2
  3. import urlparse
  4. from datetime import date
  5. # Use this when we have python 2.5
  6. #import datetime
  7. import re
  8. from BeautifulSoup import BeautifulSoup
  9. # Adding this to try to help Surrey Heath - Duncan 14/9/2007
  10. import cookielib
  11. cookie_jar = cookielib.CookieJar()
  12. ################
  13. import MultipartPostHandler
  14. # this is not mine, or part of standard python (though it should be!)
  15. # it comes from http://pipe.scs.fsu.edu/PostHandler/MultipartPostHandler.py
  16. from PlanningUtils import getPostcodeFromText, PlanningAuthorityResults, PlanningApplication
  17. date_format = "%d/%m/%Y"
  18. #This is to get the system key out of the info url
  19. system_key_regex = re.compile("TheSystemkey=(\d*)", re.IGNORECASE)
  20. # We allow the optional > for Bridgnorth, which doesn't have broken html
  21. end_head_regex = re.compile("</head>?", re.IGNORECASE)
  22. class AcolnetParser:
  23. received_date_format = "%d/%m/%Y"
  24. comment_qs_template = "ACTION=UNWRAP&RIPNAME=Root.PgeCommentForm&TheSystemkey=%s"
  25. # There is no online comment facility in these, so we provide an
  26. # appropriate email address instead
  27. comments_email_address = None
  28. # The optional amp; is to cope with Oldham, which seems to have started
  29. # quoting this url.
  30. action_regex = re.compile("<form[^>]*action=\"([^\"]*ACTION=UNWRAP&(?:amp;)?RIPSESSION=[^\"]*)\"[^>]*>", re.IGNORECASE)
  31. def _getResultsSections(self, soup):
  32. """In most cases, there is a table per app."""
  33. return soup.findAll("table", {"class": "results-table"})
  34. def _getCouncilReference(self, app_table):
  35. # return app_table.findAll("a")[1].string.strip()
  36. return app_table.a.string.strip()
  37. def _getDateReceived(self, app_table):
  38. date_str = ''.join(app_table.find(text="Registration Date:").findNext("td").string.strip().split())
  39. day, month, year = date_str.split('/')
  40. return date(int(year), int(month), int(day))
  41. # This will be better from python 2.5
  42. #return datetime.datetime.strptime(date_str, self.received_date_format)
  43. def _getAddress(self, app_table):
  44. return app_table.find(text="Location:").findNext("td").string.strip()
  45. def _getDescription(self, app_table):
  46. return app_table.find(text="Proposal:").findNext("td").string.strip()
  47. def _getInfoUrl(self, app_table):
  48. """Returns the info url for this app.
  49. We also set the system key on self._current_application,
  50. as we'll need that for the comment url.
  51. """
  52. url = app_table.a['href']
  53. self._current_application.system_key = system_key_regex.search(url).groups()[0]
  54. return urlparse.urljoin(self.base_url, url)
  55. def _getCommentUrl(self, app_table):
  56. """This must be run after _getInfoUrl"""
  57. if self.comments_email_address:
  58. return self.comments_email_address
  59. split_info_url = urlparse.urlsplit(self._current_application.info_url)
  60. comment_qs = self.comment_qs_template %self._current_application.system_key
  61. return urlparse.urlunsplit(split_info_url[:3] + (comment_qs,) + split_info_url[4:])
  62. def __init__(self,
  63. authority_name,
  64. authority_short_name,
  65. base_url,
  66. debug=False):
  67. self.authority_name = authority_name
  68. self.authority_short_name = authority_short_name
  69. self.base_url = base_url
  70. self.debug = debug
  71. # This in where we store the results
  72. self._results = PlanningAuthorityResults(self.authority_name, self.authority_short_name)
  73. # This will store the planning application we are currently working on.
  74. self._current_application = None
  75. def _cleanupHTML(self, html):
  76. """This method should be overridden in subclasses to perform site specific
  77. HTML cleanup."""
  78. return html
  79. def _getSearchResponse(self):
  80. # It looks like we sometimes need to do some stuff to get around a
  81. # javascript redirect and cookies.
  82. search_form_request = urllib2.Request(self.base_url)
  83. search_form_response = urllib2.urlopen(search_form_request)
  84. return search_form_response
  85. def getResultsByDayMonthYear(self, day, month, year):
  86. # first we fetch the search page to get ourselves some session info...
  87. search_form_response = self._getSearchResponse()
  88. search_form_contents = search_form_response.read()
  89. # This sometimes causes a problem in HTMLParser, so let's just get the link
  90. # out with a regex...
  91. groups = self.action_regex.search(search_form_contents).groups()
  92. action = groups[0]
  93. #print action
  94. # This is to handle the amp; which seems to have appeared in this
  95. # url on the Oldham site
  96. action = ''.join(action.split('amp;'))
  97. action_url = urlparse.urljoin(self.base_url, action)
  98. #print action_url
  99. our_date = date(year, month, day)
  100. search_data = {"regdate1": our_date.strftime(date_format),
  101. "regdate2": our_date.strftime(date_format),
  102. }
  103. opener = urllib2.build_opener(MultipartPostHandler.MultipartPostHandler)
  104. response = opener.open(action_url, search_data)
  105. results_html = response.read()
  106. # This is for doing site specific html cleanup
  107. results_html = self._cleanupHTML(results_html)
  108. #some javascript garbage in the header upsets HTMLParser,
  109. #so we'll just have the body
  110. just_body = "<html>" + end_head_regex.split(results_html)[-1]
  111. #self.feed(just_body)
  112. soup = BeautifulSoup(just_body)
  113. # Each app is in a table of it's own.
  114. results_tables = self._getResultsSections(soup)
  115. for app_table in results_tables:
  116. self._current_application = PlanningApplication()
  117. self._current_application.council_reference = self._getCouncilReference(app_table)
  118. self._current_application.address = self._getAddress(app_table)
  119. # Get the postcode from the address
  120. self._current_application.postcode = getPostcodeFromText(self._current_application.address)
  121. self._current_application.description = self._getDescription(app_table)
  122. self._current_application.info_url = self._getInfoUrl(app_table)
  123. self._current_application.comment_url = self._getCommentUrl(app_table)
  124. self._current_application.date_received = self._getDateReceived(app_table)
  125. self._results.addApplication(self._current_application)
  126. return self._results
  127. def getResults(self, day, month, year):
  128. return self.getResultsByDayMonthYear(int(day), int(month), int(year)).displayXML()
  129. class BridgnorthParser(AcolnetParser):
  130. def _getResultsSections(self, soup):
  131. return soup.findAll("table", {"class": "app"})
  132. def _getCouncilReference(self, app_table):
  133. return app_table.a.string.split()[-1]
  134. def _getCommentUrl(self, app_table):
  135. """This must be run after _getInfoUrl"""
  136. #http://www2.bridgnorth-dc.gov.uk/planning/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.PgeCommentForm&TheSystemkey=46958
  137. return self._current_application.info_url.replace("NewPages", "PgeCommentForm")
  138. # Cambridgeshire, although an Acolnet site, is so different that it
  139. # may as well be handled completely separately.
  140. class CanterburyParser(AcolnetParser):
  141. """Here the apps are one row each in a big table."""
  142. def _getResultsSections(self, soup):
  143. return soup.find("table").findAll("tr")[1:]
  144. def _getDateReceived(self, app_table):
  145. date_str = app_table.findAll("td")[3].string.strip()
  146. day, month, year = date_str.split('/')
  147. return date(int(year), int(month), int(day))
  148. # This will be better once we have python 2.5
  149. #return datetime.datetime.strptime(date_str, self.received_date_format)
  150. def _getAddress(self, app_table):
  151. return app_table.findAll("td")[1].string.strip()
  152. def _getDescription(self, app_table):
  153. return app_table.findAll("td")[2].string.strip()
  154. #Kensington and chelsea is sufficiently different, it may as well be handled separately
  155. # Mid Bedfordshire - there is an acolnet here, but you have to have a username
  156. # and password to access it!
  157. class OldhamParser(AcolnetParser):
  158. def _cleanupHTML(self, html):
  159. """There is a bad table end tag in this one.
  160. Fix it before we start"""
  161. bad_table_end = '</table summary="Copyright">'
  162. good_table_end = '</table>'
  163. return html.replace(bad_table_end, good_table_end)
  164. class SouthwarkParser(AcolnetParser):
  165. def _getDateReceived(self, app_table):
  166. date_str = ''.join(app_table.find(text="Statutory start date:").findNext("td").string.strip().split())
  167. day, month, year = date_str.split('/')
  168. return date(int(year), int(month), int(day))
  169. # Use this once we have python 2.5
  170. #return datetime.datetime.strptime(date_str, self.received_date_format)
  171. class SurreyHeathParser(AcolnetParser):
  172. # This is not working yet.
  173. # _getSearchResponse is an attempt to work around
  174. # cookies and a javascript redirect.
  175. # I may have a bit more of a go at this at some point if I have time.
  176. case_number_tr = 1 # this one can be got by the td class attribute
  177. reg_date_tr = 2
  178. location_tr = 4
  179. proposal_tr = 5
  180. comments_email_address = "development-control@surreyheath.gov.uk"
  181. def _getSearchResponse(self):
  182. # It looks like we sometimes need to do some stuff to get around a
  183. # javascript redirect and cookies.
  184. search_form_request = urllib2.Request(self.base_url)
  185. # Lying about the user-agent doesn't seem to help.
  186. #search_form_request.add_header("user-agent", "Mozilla/5.0 (compatible; Konqu...L/3.5.6 (like Gecko) (Kubuntu)")
  187. search_form_response = urllib2.urlopen(search_form_request)
  188. cookie_jar.extract_cookies(search_form_response, search_form_request)
  189. print search_form_response.geturl()
  190. print search_form_response.info()
  191. print search_form_response.read()
  192. # validate_url = "https://www.public.surreyheath-online.gov.uk/whalecom7cace3215643e22bb7b0b8cc97a7/whalecom0/InternalSite/Validate.asp"
  193. # javascript_redirect_url = urlparse.urljoin(self.base_url, "/whalecom7cace3215643e22bb7b0b8cc97a7/whalecom0/InternalSite/RedirectToOrigURL.asp?site_name=public&secure=1")
  194. # javascript_redirect_request = urllib2.Request(javascript_redirect_url)
  195. # javascript_redirect_request.add_header('Referer', validate_url)
  196. # cookie_jar.add_cookie_header(javascript_redirect_request)
  197. # javascript_redirect_response = urllib2.urlopen(javascript_redirect_request)
  198. # return javascript_redirect_response
  199. # Wychavon is rather different, and will need some thought. There is no
  200. # advanced search page
  201. class BoltonLikeParser(AcolnetParser):
  202. """Note that Bolton has ceased to be BoltonLike with its latest change of url."""
  203. def _getCouncilReference(self, app_table):
  204. return app_table.findAll("a")[1].string.strip()
  205. class LewishamParser(BoltonLikeParser):
  206. comments_email_address = "planning@lewisham.com"
  207. class BassetlawParser(BoltonLikeParser):
  208. comments_email_address = "planning@bassetlaw.gov.uk"
  209. def _cleanupHTML(self, html):
  210. """There is a broken div in this page. We don't need any divs, so
  211. let's get rid of them all."""
  212. div_regex = re.compile("</?div[^>]*>", re.IGNORECASE)
  213. return div_regex.sub('', html)
  214. class HarlowParser(AcolnetParser):
  215. def _getCommentUrl(self, app_table):
  216. """This must be run after _getInfoUrl"""
  217. #http://www2.bridgnorth-dc.gov.uk/planning/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.PgeCommentForm&TheSystemkey=46958
  218. return self._current_application.info_url.replace("PgeResultDetail", "PgeCommentNeighbourForm&amp;hasreference=no")
  219. if __name__ == '__main__':
  220. day = 4
  221. month = 4
  222. year = 2008
  223. #parser = AcolnetParser("Babergh", "Babergh", "http://planning.babergh.gov.uk/dcdatav2//acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  224. #parser = AcolnetParser("Basingstoke", "Basingstoke", "http://planning.basingstoke.gov.uk/DCOnline2/acolnetcgi.exe?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  225. #parser = BassetlawParser("Bassetlaw", "Bassetlaw", "http://www.bassetlaw.gov.uk/planning/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  226. #parser = AcolnetParser("Bolton", "Bolton", "http://www.planning.bolton.gov.uk/DCOnlineV2/acolnetcgi.exe?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  227. #parser = BridgnorthParser("Bridgnorth", "Bridgnorth", "http://www2.bridgnorth-dc.gov.uk/planning/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.PgeSearch")
  228. #parser = AcolnetParser("Bury", "Bury", "http://e-planning.bury.gov.uk/DCWebPages/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  229. #parser = CanterburyParser("Canterbury", "Canterbury", "http://planning.canterbury.gov.uk/scripts/acolnetcgi.exe?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  230. parser = BoltonLikeParser("Carlisle", "Carlisle", "http://planning.carlisle.gov.uk/acolnet/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  231. #parser = AcolnetParser("Croydon", "Croydon", "http://planning.croydon.gov.uk/DCWebPages/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  232. #parser = AcolnetParser("Derby", "Derby", "http://eplanning.derby.gov.uk/acolnet/planningpages02/acolnetcgi.exe?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  233. #parser = AcolnetParser("East Lindsey", "East Lindsey", "http://www.e-lindsey.gov.uk/planning/AcolnetCGI.exe?ACTION=UNWRAP&RIPNAME=Root.pgesearch", "AcolnetParser")
  234. #parser = AcolnetParser("Exeter City Council", "Exeter", "http://pub.exeter.gov.uk/scripts/Acolnet/dataonlineplanning/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  235. #parser = BoltonParser("Fylde", "Fylde", "http://www2.fylde.gov.uk/planning/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  236. #parser = AcolnetParser("Guildford", "Guildford", "http://www.guildford.gov.uk/DLDC_Version_2/acolnetcgi.exe?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  237. #parser = AcolnetParser("Harlow", "Harlow", "http://planning.harlow.gov.uk/DLDC_Version_2/acolnetcgi.exe?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  238. #parser = AcolnetParser("Havant", "Havant", "http://www3.havant.gov.uk/scripts/planningpages/acolnetcgi.exe?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  239. #parser = BoltonLikeParser("Hertsmere", "Hertsmere", "http://www2.hertsmere.gov.uk/ACOLNET/DCOnline//acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  240. #parser = LewishamParser("Lewisham", "Lewisham", "http://acolnet.lewisham.gov.uk/lewis-xslpagesdc/acolnetcgi.exe?ACTION=UNWRAP&RIPNAME=Root.PgeSearch")
  241. #parser = AcolnetParser("Mid Suffolk", "Mid Suffolk", "http://planning.midsuffolk.gov.uk/planning/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  242. #parser = BoltonLikeParser("New Forest District Council", "New Forest DC", "http://web3.newforest.gov.uk/planningonline/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  243. #parser = BoltonLikeParser("New Forest National Park Authority", "New Forest NPA", "http://web01.newforestnpa.gov.uk/planningpages/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  244. #parser = AcolnetParser("North Hertfordshire", "North Herts", "http://www.north-herts.gov.uk/dcdataonline/Pages/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.PgeSearch")
  245. #parser = AcolnetParser("North Wiltshire", "North Wilts", "http://planning.northwilts.gov.uk/DCOnline/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  246. #parser = OldhamParser("Oldham", "Oldham", "http://planning.oldham.gov.uk/planning/AcolNetCGI.gov?ACTION=UNWRAP&Root=PgeSearch")
  247. #parser = BoltonLikeParser("Renfrewshire", "Renfrewshire", "http://planning.renfrewshire.gov.uk/acolnetDCpages/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.PgeSearch")
  248. #parser = AcolnetParser("South Bedfordshire", "South Bedfordshire", "http://planning.southbeds.gov.uk/plantech/DCWebPages/acolnetcgi.exe?ACTION=UNWRAP&RIPNAME=Root.PgeSearch")
  249. #parser = SouthwarkParser("London Borough of Southwark", "Southwark", "http://planningonline.southwarksites.com/planningonline2/AcolNetCGI.exe?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  250. #parser = AcolnetParser("Suffolk Coastal", "Suffolk Coastal", "http://apps3.suffolkcoastal.gov.uk/DCDataV2/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  251. #parser = AcolnetParser("Surrey Heath", "Surrey Heath", "https://www.public.surreyheath-online.gov.uk/whalecom60b1ef305f59f921/whalecom0/Scripts/PlanningPagesOnline/acolnetcgi.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  252. parser = AcolnetParser("Stockport Metropolitan Borough Council", "Stockport", "http://planning.stockport.gov.uk/PlanningData/AcolNetCGI.gov?ACTION=UNWRAP&RIPNAME=Root.pgesearch")
  253. print parser.getResults(day, month, year)