adrianshort
/
planningalerts
kopie van https://github.com/adrianshort/planningalerts.git
1
0
Vork 0
Code Kwesties 0 Publicaties 0 Wiki Activiteit
Automatically exported from code.google.com/p/planningalerts
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
177 Commits
2 Branches
1.3 MiB
 
 
 
 
 
 
Tree: 7e423683f5
Branches Labels
${ item.name }
Maak branch ${ searchTerm }
van '7e423683f5'
${ noResults }
planningalerts/python_scrapers/SwiftLG.py

227 regels
10 KiB
Ruw Blame Geschiedenis 

  1. 

  2. import urllib2

  3. import urllib

  4. import urlparse

  5. import cgi

  6. import re

  7. import datetime

  8. 

  9. from BeautifulSoup import BeautifulSoup

  10. 

  11. from PlanningUtils import getPostcodeFromText, \

  12.     PlanningAuthorityResults, \

  13.     PlanningApplication

  14. 

  15. # - Browser request: --------------------------

  16. # {POST http://digitalmaidstone.co.uk/swiftlg/apas/run/WPHAPPCRITERIA HTTP/1.0} {Host: digitalmaidstone.co.uk

  17. # Accept: text/html, text/plain, text/css, text/sgml, */*;q=0.01

  18. # Accept-Encoding: gzip

  19. # Accept-Language: en

  20. # Pragma: no-cache

  21. # Cache-Control: no-cache

  22. # User-Agent: Lynx/2.8.6rel.4 libwww-FM/2.14 SSL-MM/1.4.1 GNUTLS/1.6.3

  23. # Content-type: application/x-www-form-urlencoded

  24. # Content-length: 638

  25. # } %25.MAINBODY.WPACIS.1.=&APNID.MAINBODY.WPACIS.1.=&JUSTLOCATION.MAINBODY.WPACIS.1.=&JUSTDEVDESC.MAINBODY.WPACIS.1.=&DEVDESC.MAINBODY.WPACIS.1.=&SURNAME.MAINBODY.WPACIS.1.=&REGFROMDATE.MAINBODY.WPACIS.1.=01%2F11%2F2007&REGTODATE.MAINBODY.WPACIS.1.=02%2F11%2F2007&DECFROMDATE.MAINBODY.WPACIS.1.=&DECTODATE.MAINBODY.WPACIS.1.=&FINALGRANTFROM.MAINBODY.WPACIS.1.=&FINALGRANTTO.MAINBODY.WPACIS.1.=&APELDGDATFROM.MAINBODY.WPACIS.1.=&APELDGDATTO.MAINBODY.WPACIS.1.=&APEDECDATFROM.MAINBODY.WPACIS.1.=&APEDECDATTO.MAINBODY.WPACIS.1.=&AREA.MAINBODY.WPACIS.1.=&WARD.MAINBODY.WPACIS.1.=&PARISH.MAINBODY.WPACIS.1.=&SEARCHBUTTON.MAINBODY.WPACIS.1.=Search

  26. # server=[digitalmaidstone.co.uk] , port=[80], script=[/swiftlg/apas/run/WPHAPPCRITERIA]

  27. # request_line=[POST /swiftlg/apas/run/WPHAPPCRITERIA HTTP/1.0]

  28. 

  29. # second page

  30. #http://digitalmaidstone.co.uk/swiftlg/apas/run/WPHAPPSEARCHRES.displayResultsURL?ResultID=243941&

  31. #StartIndex=11&

  32. #SortOrder=APNID:asc&

  33. #DispResultsAs=WPHAPPSEARCHRES&

  34. #BackURL=<a%20href=wphappcriteria.display?paSearchKey=147118>Search%20Criteria

  35. 

  36. # Date format to enter into search boxes

  37. date_format = "%d/%m/%Y"

  38. 

  39. class SwiftLGParser:

  40.     search_path = "WPHAPPCRITERIA"

  41.     info_path = "WPHAPPDETAIL.DisplayUrl?theApnID=%s"

  42.     comment_path ="wphmakerep.displayURL?ApnID=%s"

  43. 

  44.     def _findResultsTable(self, soup):

  45.         """Unless there is just one table in the page, the resuts table,

  46.         override this in a subclass."""

  47.         return soup.table

  48. 

  49.     def _findTRs(self, results_table):

  50.         """The usual situation is for the results table to contain

  51.         one row of headers, followed by a row per app.

  52.         If this is not the case, override this in a subclass."""

  53.         return results_table.findAll("tr")[1:]

  54. 

  55.     def __init__(self,

  56.                  authority_name,

  57.                  authority_short_name,

  58.                  base_url,

  59.                  debug=False):

  60.         

  61.         self.authority_name = authority_name

  62.         self.authority_short_name = authority_short_name

  63.         self.base_url = base_url

  64. 

  65.         self.search_url = urlparse.urljoin(base_url, self.search_path)

  66.         self.info_url = urlparse.urljoin(base_url, self.info_path)

  67.         self.comment_url = urlparse.urljoin(base_url, self.comment_path)

  68. 

  69.         self.debug = debug

  70. 

  71.         self._results = PlanningAuthorityResults(self.authority_name, self.authority_short_name)

  72. 

  73. 

  74.     def getResultsByDayMonthYear(self, day, month, year):

  75.         search_date = datetime.date(year, month, day)

  76.         

  77.         post_data = urllib.urlencode((

  78.                 ("REGFROMDATE.MAINBODY.WPACIS.1.", search_date.strftime(date_format)),

  79.                 ("REGTODATE.MAINBODY.WPACIS.1.", search_date.strftime(date_format)),

  80.                 ("SEARCHBUTTON.MAINBODY.WPACIS.1.", "Search"),

  81.                 ))

  82.         

  83.         response = urllib2.urlopen(self.search_url, post_data)

  84.         contents = response.read()

  85. 

  86.         # Check for the no results warning

  87.         if not contents.count("No Matching Applications Found"):

  88.             soup = BeautifulSoup(contents)

  89. 

  90.             # Get the links to later pages of results.

  91.             later_pages = soup.findAll("a", {"href": re.compile("WPHAPPSEARCHRES\.displayResultsURL.*StartIndex=\d*.*")})

  92. 

  93.             for a in ["initial_search"] + later_pages:

  94.                 if a != "initial_search":

  95.                     url = a['href']

  96. 

  97.                     # Example url

  98. 

  99.                     #http://digitalmaidstone.co.uk/swiftlg/apas/run/WPHAPPSEARCHRES.displayResultsURL?ResultID=244037&StartIndex=11&SortOrder=APNID:asc&DispResultsAs=WPHAPPSEARCHRES&BackURL=<a href=wphappcriteria.display?paSearchKey=147170>Search Criteria</a>

  100. 

  101.                     # urllib2 doesn't like this url, to make it happy, we'll

  102.                     # get rid of the BackURL parameter, which we don't need.

  103. 

  104.                     split_url = urlparse.urlsplit(url)

  105.                     qs = split_url[3]

  106. 

  107.                     # This gets us a dictionary of key to lists of values

  108.                     qsl = cgi.parse_qsl(qs)

  109. 

  110.                     # Get rid of BackURL

  111.                     qsl.pop(-1)

  112. 

  113.                     # I think this is safe, as there are no repeats of parameters

  114.                     new_qs = urllib.urlencode(qsl)

  115. 

  116.                     url = urlparse.urlunsplit(split_url[:3] + (new_qs,) + split_url[4:])

  117. 

  118.                     this_page_url = urlparse.urljoin(self.base_url, url)

  119.                     response = urllib2.urlopen(this_page_url)

  120.                     contents = response.read()

  121.                     soup = BeautifulSoup(contents)

  122. 

  123.                 results_table = self._findResultsTable(soup)#.body.find("table", {"class": "apas_tbl"})

  124. 

  125.                 trs = self._findTRs(results_table)

  126. 

  127.                 for tr in trs:

  128.                     self._current_application = PlanningApplication()

  129. 

  130.                     tds = tr.findAll("td")

  131. 

  132.                     # The first td

  133. 

  134.                     #<td class="apas_tblContent"><a href="WPHAPPDETAIL.DisplayUrl?theApnID=07/1884&amp;backURL=&lt;a href=wphappcriteria.display?paSearchKey=147125&gt;Search Criteria&lt;/a&gt; &gt; &lt;a href='wphappsearchres.displayResultsURL?ResultID=243950%26StartIndex=1%26SortOrder=APNID:asc%26DispResultsAs=WPHAPPSEARCHRES%26BackURL=&lt;a href=wphappcriteria.display?paSearchKey=147125&gt;Search Criteria&lt;/a&gt;'&gt;Search Results&lt;/a&gt;"></a><a href="wphappcriteria.display?paSearchKey=147125">Search Criteria</a> > <a href="wphappsearchres.displayResultsURL?ResultID=243950%26StartIndex=1%26SortOrder=APNID:asc%26DispResultsAs=WPHAPPSEARCHRES%26BackURL=&lt;a href=wphappcriteria.display?paSearchKey=147125&gt;Search Criteria&lt;/a&gt;"></a><a href="wphappcriteria.display?paSearchKey=147125">Search Criteria</a>'>Search Results">07/1884</td>

  135. 

  136.                     # The html here is a bit of a mess, and doesn't all get into

  137.                     # the soup.

  138.                     # We can get the reference from the first <a href> in td 0.

  139.                     first_link = tds[0].a['href']

  140. 

  141.                     app_id = cgi.parse_qs(urlparse.urlsplit(first_link)[3])['theApnID'][0]

  142. 

  143.                     self._current_application.date_received = search_date

  144.                     self._current_application.council_reference = app_id

  145.                     self._current_application.info_url = self.info_url %(app_id)

  146.                     self._current_application.comment_url = self.comment_url %(app_id)

  147.                     self._current_application.description = tds[1].string.strip()

  148. 

  149.                     # the second td

  150. 

  151.                     #<td class="apas_tblContent"><input type="HIDDEN" name="ORDERCOUNTER.PAHEADER.PACIS2.1-1." value="1" class="input-box" size="7" />

  152.                     #LAND ADJ. BRAMBLING, HAWKENBURY ROAD, HAWKENBURY, TN120EA

  153.                     #</td>

  154. 

  155.                     # For some reason, this doesn't work:

  156.                     #address = tds[2].string

  157. 

  158.                     # But this does

  159.                     address = tds[2].input.next.strip()

  160. 

  161.                     self._current_application.address = address

  162.                     self._current_application.postcode = getPostcodeFromText(address)

  163. 

  164.                     self._results.addApplication(self._current_application)

  165. 

  166.         return self._results

  167. 

  168.     def getResults(self, day, month, year):

  169.         return self.getResultsByDayMonthYear(int(day), int(month), int(year)).displayXML()

  170. 

  171. class EastHertsParser(SwiftLGParser):

  172.     def _findResultsTable(self, soup):

  173.         return soup.findAll("table")[3]

  174. 

  175. class GwyneddParser(SwiftLGParser):

  176.     def _findResultsTable(self, soup):

  177.         return soup.find("table", {"class": "thinBox"})

  178. 

  179. class IslingtonParser(SwiftLGParser):

  180.     def _findResultsTable(self, soup):

  181.         return soup.table.table

  182. 

  183. class LakeDistrictParser(SwiftLGParser):

  184.     def _findResultsTable(self, soup):

  185.         return soup.table.table

  186. 

  187. class MacclesfieldParser(SwiftLGParser):

  188.     def _findResultsTable(self, soup):

  189.         return soup.findAll("table")[6]

  190. 

  191. class MoleValleyParser(SwiftLGParser):

  192.     def _findResultsTable(self, soup):

  193.         return soup.findAll("table")[5]

  194. 

  195. class SloughParser(SwiftLGParser):

  196.     def _findResultsTable(self, soup):

  197.         return soup.findAll("table")[1]

  198. 

  199.     def _findTRs(self, results_table):

  200.         return results_table.findAll("tr")[2:]

  201. 

  202. if __name__ == '__main__':

  203.     parser = SwiftLGParser("Boston Borough Council", "Boston", "http://195.224.121.199/swiftlg/apas/run/")

  204. #    parser = SwiftLGParser("Dudley", "Dudley", "http://www2.dudley.gov.uk/swiftlg/apas/run/")

  205. #    parser = EastHertsParser("East Hertfordshire", "East Herts", "http://e-services.eastherts.gov.uk/swiftlg/apas/run/")

  206. #    parser = GwyneddParser("Gwynedd", "Gwynedd", "http://www.gwynedd.gov.uk/swiftlg/apas/run/")

  207. #    parser = IslingtonParser("Islington", "Islington", "https://www.islington.gov.uk/onlineplanning/apas/run/")

  208. #    parser = LakeDistrictParser("Lake District", "Lake District", "http://www.lake-district.gov.uk/swiftlg/apas/run/")

  209. #    parser = SwiftLGParser("Maidstone Borough Council", "Maidstone", "http://digitalmaidstone.co.uk/swiftlg/apas/run/")

  210. #    parser = MoleValleyParser("Mole Valley", "Mole Valley", "http://www.molevalley.gov.uk/swiftlg/apas/run/")

  211. #    parser = SwiftLGParser("Pembrokeshire County Council", "Pembrokeshire", "http://planning.pembrokeshire.gov.uk/swiftlg/apas/run/")

  212. #    parser = SwiftLGParser("Rochdale Metropolitan Borough Council", "Rochdale", "http://www.rochdale.gov.uk/swiftlg/apas/run/")

  213. #    parser = SloughParser("Slough", "Slough", "http://www2.slough.gov.uk/swiftlg/apas/run/")

  214. #    parser = SwiftLGParser("Snowdonia National Park", "Snowdonia", "http://www.snowdonia-npa.gov.uk/swiftlg/apas/run/")

  215. #    parser = SwiftLGParser("St Edmundsbury", "Bury St Edmunds", "http://www.stedmundsbury.gov.uk/swiftlg/apas/run/")

  216. #    parser = MacclesfieldParser("Macclesfield", "Macclesfield", "http://www.planportal.macclesfield.gov.uk/swiftlg/apas/run/")

  217.     print parser.getResults(22,11,2007)

  218. 

  219. 

  220. # To Do:

  221. 

  222. #1) Check out comment url on Maidstone

  223. 

  224. #2) Daventry, when it is back up.

  225. 

  226. #3) Work out what goes wrong with Gwynedd on 06/11/2007