more scrapers
[disclosr.git] / documents / genericScrapers.py
blob:a/documents/genericScrapers.py -> blob:b/documents/genericScrapers.py
--- a/documents/genericScrapers.py
+++ b/documents/genericScrapers.py
@@ -9,6 +9,7 @@
 import dateutil
 from dateutil.parser import *
 from datetime import *
+import codecs
 
 class GenericDisclogScraper(object):
         __metaclass__ = abc.ABCMeta
@@ -84,12 +85,18 @@
                 doc.update({'description': descriptiontxt})
 		return
         def getTitle(self, content, entry, doc):
-                doc.update({'title': content.string})
+                doc.update({'title': (''.join(content.stripped_strings))})
 		return
 	def getTable(self, soup):
 		return soup.table
+	def getRows(self, table):
+		return table.find_all('tr')
 	def getDate(self, content, entry, doc):
-		edate = parse(''.join(content.stripped_strings).strip(), dayfirst=True, fuzzy=True).strftime("%Y-%m-%d")
+		date = ''.join(content.stripped_strings).strip()
+		(a,b,c) = date.partition("(")
+		date = self.remove_control_chars(a.replace("Octber","October"))
+		print date
+		edate = parse(date, dayfirst=True, fuzzy=True).strftime("%Y-%m-%d")
 		print edate
 		doc.update({'date': edate})
 		return
@@ -110,11 +117,11 @@
 			# http://www.crummy.com/software/BeautifulSoup/documentation.html
 				soup = BeautifulSoup(content)
 				table = self.getTable(soup)
-				for row in table.find_all('tr'):
+				for row in self.getRows(table):
 					columns = row.find_all('td')
 					if len(columns) == self.getColumnCount():
-						(id, date, description, title, notes) = self.getColumns(columns)
-						print ''.join(id.stripped_strings)
+						(id, date, title, description, notes) = self.getColumns(columns)
+						print self.remove_control_chars(''.join(id.stripped_strings))
 						if id.string == None:
 							hash = scrape.mkhash(self.remove_control_chars(url+(''.join(date.stripped_strings))))
 						else:
@@ -123,14 +130,19 @@
 							
 						if doc == None:
 							print "saving " +hash
-							doc = {'_id': hash, 'agencyID': self.getAgencyID(), 'url': self.getURL(), 'docID': id.string}
+							doc = {'_id': hash, 'agencyID': self.getAgencyID(), 'url': self.getURL(), 'docID': (''.join(id.stripped_strings))}
 							self.getLinks(self.getURL(),row,doc)
                                 			self.getTitle(title,row, doc)
                                 			self.getDate(date,row, doc)
 							self.getDescription(description,row, doc)
 							if notes != None:
-                                        			doc.update({ 'notes': notes.string})
-							foidocsdb.save(doc)
+                                        			doc.update({ 'notes': (''.join(notes.stripped_strings))})
+                                                        badtitles = ['-','Summary of FOI Request','FOI request(in summary form)','Summary of FOI request received by the ASC',
+'Summary of FOI request received by agency/minister','Description of Documents Requested','FOI request','Description of FOI Request','Summary of request','Description','Summary',
+'Summary of FOIrequest received by agency/minister','Summary of FOI request received','Description of    FOI Request',"FOI request",'Results 1 to 67 of 67']
+							if doc['title'] not in badtitles and doc['description'] != '':
+                                                            print "saving"
+                                                            foidocsdb.save(doc)
 						else:
 							print "already saved "+hash