From fb7ba977ae79fa75c8435ada177c311269e843d8 Mon Sep 17 00:00:00 2001
From: "duncan.parkes" <duncan.parkes@cf68f80a-222c-0410-aabe-5f79d0504a29>
Date: Fri, 8 Aug 2008 16:50:32 +0000
Subject: [PATCH] Add Carmarthenshire scraper.

---
 python_scrapers/Carmarthenshire.py   | 78 ++++++++++++++++++++++++++++
 python_scrapers/OtherFilesToCopy.csv |  1 +
 python_scrapers/SitesToGenerate.csv  |  1 +
 3 files changed, 80 insertions(+)
 create mode 100644 python_scrapers/Carmarthenshire.py

diff --git a/python_scrapers/Carmarthenshire.py b/python_scrapers/Carmarthenshire.py
new file mode 100644
index 0000000..bd08ac5
--- /dev/null
+++ b/python_scrapers/Carmarthenshire.py
@@ -0,0 +1,78 @@
+import urllib2
+import urllib
+import urlparse
+
+import datetime, time
+import cgi
+
+from BeautifulSoup import BeautifulSoup
+
+from PlanningUtils import PlanningApplication, \
+    PlanningAuthorityResults, \
+    getPostcodeFromText
+
+class CarmarthenshireParser:
+    def __init__(self, *args):
+        self.comments_email_address = "planning@carmarthenshire.gov.uk"
+
+        self.authority_name = "Carmarthenshire County Council"
+        self.authority_short_name = "Carmarthenshire"
+        self.base_url = "http://www.carmarthenshire.gov.uk/CCC_APPS/eng/plannaps/CCC_PlanningApplicationsResults.asp?datemode=range&in_lo_date=%(day)s%%2F%(month)s%%2F%(year)s&in_hi_date=%(day)s%%2F%(month)s%%2F%(year)s&SUBMIT=Search"
+
+        self._results = PlanningAuthorityResults(self.authority_name, self.authority_short_name)
+
+
+    def getResultsByDayMonthYear(self, day, month, year):
+        search_day = datetime.date(year, month, day)
+
+        # Now get the search page
+        response = urllib2.urlopen(self.base_url %{"day": day,
+                                                   "month": month,
+                                                   "year": year,
+                                                   })
+        soup = BeautifulSoup(response.read())
+
+        trs = soup.findAll("tr", valign="middle")
+
+        count = 0
+        for tr in trs:
+            # The odd trs are just spacers
+            if count % 2 == 0:
+                application = PlanningApplication()
+
+                tds = tr.findAll("td")
+                
+                application.date_received = search_day
+                application.council_reference = tds[1].a.string
+                application.address = tds[3].a.string
+                application.postcode = getPostcodeFromText(application.address)
+                
+                # All the links in this <tr> go to the same place...
+                application.info_url = urlparse.urljoin(self.base_url, tr.a['href'])
+
+                # Still looking for description and comment url
+                
+                # For the description, we'll need the info page
+                info_soup = BeautifulSoup(urllib2.urlopen(application.info_url).read())
+
+                application.description = info_soup.find(text="Description").findNext("td").findNext("td").font.string
+
+                # While we're here, lets get the OSGB grid ref
+                application.osgb_x, application.osgb_y = info_soup.find(text="Grid Reference").findNext("td").font.string.split("-")
+
+                # We'll have to use an email address for comments
+                application.comment_url = self.comments_email_address
+
+                self._results.addApplication(application)
+
+            count += 1
+
+        return self._results
+
+    def getResults(self, day, month, year):
+        return self.getResultsByDayMonthYear(int(day), int(month), int(year)).displayXML()
+
+if __name__ == '__main__':
+    parser = CarmarthenshireParser()
+    print parser.getResults(8,8,2008)
+
diff --git a/python_scrapers/OtherFilesToCopy.csv b/python_scrapers/OtherFilesToCopy.csv
index d47dc1d..e03987c 100644
--- a/python_scrapers/OtherFilesToCopy.csv
+++ b/python_scrapers/OtherFilesToCopy.csv
@@ -45,3 +45,4 @@
 "AmberValley.py", "420"
 "Aberdeenshire.py", "420"
 "Brent.py", "420"
+"Carmarthenshire.py", "420"
diff --git a/python_scrapers/SitesToGenerate.csv b/python_scrapers/SitesToGenerate.csv
index fdf069f..ebc23c4 100644
--- a/python_scrapers/SitesToGenerate.csv
+++ b/python_scrapers/SitesToGenerate.csv
@@ -249,3 +249,4 @@
 "Amber Valley Borough Council", "Amber Valley", "", "AmberValley", "AmberValleyParser"
 "Aberdeenshire Council", "Aberdeenshire", "", "Aberdeenshire", "AberdeenshireParser"
 "London Borough of Brent", "Brent", "", "Brent", "BrentParser"
+"Carmarthenshire County Council", "Carmarthenshire", "", "Carmarthenshire", "CarmarthenshireParser"