fix agd scraper
fix agd scraper


Former-commit-id: 387571086ee1f51b4ec59e3de6959dfb4f78beff

--- a/documents/genericScrapers.py
+++ b/documents/genericScrapers.py
@@ -7,11 +7,14 @@
 from datetime import datetime
 import feedparser
 import abc
+import unicodedata, re
 
 class GenericDisclogScraper(object):
         __metaclass__ = abc.ABCMeta
 	agencyID = None
 	disclogURL = None
+	def remove_control_chars(self, input):
+		return "".join([i for i in input if ord(i) in range(32, 127)])
         def getAgencyID(self):
                 """ disclosr agency id """
 		if self.agencyID == None:
@@ -79,6 +82,8 @@
                 	descriptiontxt = descriptiontxt + " \n" + string
                 doc.update({'description': descriptiontxt})
 		return
+	def getTable(self, soup):
+		return soup.table
 
 	def doScrape(self):
 		cal = pdt.Calendar()
@@ -88,12 +93,16 @@
 			if mime_type == "text/html" or mime_type == "application/xhtml+xml" or mime_type =="application/xml":
 			# http://www.crummy.com/software/BeautifulSoup/documentation.html
 				soup = BeautifulSoup(content)
-				for row in soup.table.find_all('tr'):
+				table = self.getTable(soup)
+				for row in table.find_all('tr'):
 					columns = row.find_all('td')
 					if len(columns) == self.getColumnCount():
 						(id, date, description, title, notes) = self.getColumns(columns)
 						print id.string
-						hash = scrape.mkhash(url+id.string)
+						if id.string == None:
+							hash = scrape.mkhash(self.remove_control_chars(url+date.string))
+						else:
+							hash = scrape.mkhash(self.remove_control_chars(url+id.string))
 						links = []
 						for atag in row.find_all("a"):
 							if atag.has_key('href'):
@@ -108,13 +117,12 @@
 								print dtdate
                                                         	edate = ""+str(dtdate[0])+'-'+str(dtdate[1])+'-'+str(dtdate[2])
 							else:
-								edate = datetime.strptime(date.string, "%d %B %Y").strftime("%Y-%m-%d")
+								edate = datetime.strptime(date.string.strip(), "%d %B %Y").strftime("%Y-%m-%d")
 							doc = {'_id': hash, 'agencyID': self.getAgencyID(), 'url': self.getURL(), 'docID': id.string,
 			 				 "date": edate,"title": title.string}
-							self.getDescription(description,row, doc)
-
                                 			if links != []:
                                         			doc.update({'links': links})
+							self.getDescription(description,row, doc)
                                 			if notes != None:
                                         			doc.update({ 'notes': notes.string})
 							foidocsdb.save(doc)

--- a/documents/scrape.py
+++ b/documents/scrape.py
@@ -82,7 +82,7 @@
 	url = canonurl(url)
 	hash = mkhash(url)
 	req = urllib2.Request(url)
-	print "Fetching %s" % url
+	print "Fetching %s (%s)" % (url,hash)
 	if url.startswith("mailto") or url.startswith("javascript") or url.startswith("#") or url == None or url == "":
 		print "Not a valid HTTP url"
 		return (None,None,None)
@@ -94,7 +94,8 @@
 			print "Uh oh, trying to scrape URL again too soon!"
 			last_attachment_fname = doc["_attachments"].keys()[-1]
 			last_attachment = docsdb.get_attachment(doc,last_attachment_fname)
-			return (doc['url'],doc['mime_type'],last_attachment.read())
+			content = last_attachment
+			return (doc['url'],doc['mime_type'],content)
 		if scrape_again == False:
 			print "Not scraping this URL again as requested"
 			return (None,None,None)

--- a/documents/scrapers/227cb6eb7d2c9f8a6e846df7447d6caa.py
+++ b/documents/scrapers/227cb6eb7d2c9f8a6e846df7447d6caa.py
@@ -1,42 +1,42 @@
 import sys,os
 sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../'))
 import genericScrapers
-#RSS feed not detailed
+import scrape
+from bs4 import BeautifulSoup
 
 #http://www.doughellmann.com/PyMOTW/abc/
 class ScraperImplementation(genericScrapers.GenericOAICDisclogScraper):
         def getDescription(self,content, entry,doc):
 		link = None
+                links = []
+                description = ""
 		for atag in entry.find_all('a'):
 			if atag.has_key('href'):
-				link = scrape.fullurl(url,atag['href'])			
-                (url,mime_type,htcontent) = scrape.fetchURL(scrape.docsdb, link, "foidocuments", self.getAgencyID(), False)
-                if htcontent != None:
-                        if mime_type == "text/html" or mime_type == "application/xhtml+xml" or mime_type =="application/xml":
-                        # http://www.crummy.com/software/BeautifulSoup/documentation.html
-                                soup = BeautifulSoup(content)
-                                links = []
-                                description = ""
-                                dldivs = soup.find('div',class_="download")
-                                if dldivs != None:
-                                        for atag in dldivs.find_all("a"):
-                                                if atag.has_key('href'):
-                                                        links.append(scrape.fullurl(url,atag['href']))
-                                nodldivs = soup.find('div',class_="incompleteNotification")
-                                if nodldivs != None and nodldivs.stripped_strings != None:
-                                        for text in nodldivs.stripped_strings:
-                                                description = description + text
-                                for row in soup.table.find_all('tr'):
-                                        if row != None:
-                                                description = description + "\n" + row.find('th').string + ": "
-                                                for text in row.find('div').stripped_strings:
-                                                         description = description + text
-                                if links != []:
-                                        doc.update({'links': links})
-                                if description != "":
-                                        doc.update({ 'description': description})
+				link = scrape.fullurl(self.getURL(),atag['href'])			
+                                (url,mime_type,htcontent) = scrape.fetchURL(scrape.docsdb, link, "foidocuments", self.getAgencyID(), False)
+                                if htcontent != None:
+                                        if mime_type == "text/html" or mime_type == "application/xhtml+xml" or mime_type =="application/xml":
+                                        # http://www.crummy.com/software/BeautifulSoup/documentation.html
+                                                soup = BeautifulSoup(htcontent)
+                                                for row in soup.find(class_ = "ms-rteTable-GreyAlternating").find_all('tr'):
+                                                        if row != None:
+								rowtitle = row.find('th').string
+                                                                description = description + "\n" + rowtitle + ": "
+                                                                for text in row.find('td').stripped_strings:
+                                                                        description = description + text
+                                                     		for atag in row.find_all("a"):
+                                                                	if atag.has_key('href'):
+                                                                        	links.append(scrape.fullurl(link,atag['href']))
+
+		if links != []:
+                 	doc.update({'links': links})
+                if description != "":
+                        doc.update({ 'description': description})
+
 	def getColumnCount(self):
 		return 2
+	def getTable(self,soup):
+		return soup.find(class_ = "ms-rteTable-GreyAlternating")
 	def getColumns(self,columns):
 		(date, title) = columns
 		return (title, date, title, title, None)