pdf scrapers
Former-commit-id: 6a33167b9cf20ed9af0d41f252320532648535db
--- /dev/null
+++ b/documents/disclosr-documents.nja
@@ -1,1 +1,7 @@
-
+{
+ "venv": "",
+ "project-type": "Import from sources",
+ "name": "disclosr-documents",
+ "license": "GNU General Public License v3",
+ "description": ""
+}
--- a/documents/genericScrapers.py
+++ b/documents/genericScrapers.py
@@ -1,155 +1,255 @@
-import sys,os
+import sys
+import os
sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../'))
import scrape
from bs4 import BeautifulSoup
from time import mktime
import feedparser
import abc
-import unicodedata, re
+import unicodedata
+import re
import dateutil
from dateutil.parser import *
from datetime import *
import codecs
+from StringIO import StringIO
+
+from pdfminer.pdfparser import PDFDocument, PDFParser
+from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter, process_pdf
+from pdfminer.pdfdevice import PDFDevice, TagExtractor
+from pdfminer.converter import TextConverter
+from pdfminer.cmapdb import CMapDB
+from pdfminer.layout import LAParams
+
+
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:
- self.agencyID = os.path.basename(sys.argv[0]).replace(".py","")
- return self.agencyID
-
- def getURL(self):
- """ disclog URL"""
- if self.disclogURL == None:
- agency = scrape.agencydb.get(self.getAgencyID())
- self.disclogURL = agency['FOIDocumentsURL']
- return self.disclogURL
-
- @abc.abstractmethod
- def doScrape(self):
- """ do the scraping """
- return
-
- @abc.abstractmethod
- def getDescription(self, content, entry, doc):
- """ get description"""
- return
-
+ __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 is None:
+ self.agencyID = os.path.basename(sys.argv[0]).replace(".py", "")
+ return self.agencyID
+
+ def getURL(self):
+ """ disclog URL"""
+ if self.disclogURL is None:
+ agency = scrape.agencydb.get(self.getAgencyID())
+ self.disclogURL = agency['FOIDocumentsURL']
+ return self.disclogURL
+
+ @abc.abstractmethod
+ def doScrape(self):
+ """ do the scraping """
+ return
+
+
+class GenericPDFDisclogScraper(GenericDisclogScraper):
+
+ def doScrape(self):
+ foidocsdb = scrape.couch['disclosr-foidocuments']
+ (url, mime_type, content) = scrape.fetchURL(scrape.docsdb,
+ self.getURL(), "foidocuments", self.getAgencyID())
+ laparams = LAParams()
+ rsrcmgr = PDFResourceManager(caching=True)
+ outfp = StringIO()
+ device = TextConverter(rsrcmgr, outfp, codec='utf-8',
+ laparams=laparams)
+ fp = StringIO()
+ fp.write(content.read())
+
+ process_pdf(rsrcmgr, device, fp, set(), caching=True,
+ check_extractable=True)
+ description = outfp.getvalue()
+ fp.close()
+ device.close()
+ outfp.close()
+ dochash = scrape.mkhash(description)
+ doc = foidocsdb.get(dochash)
+ if doc is None:
+ print "saving " + dochash
+ edate = date.today().strftime("%Y-%m-%d")
+ doc = {'_id': dochash, 'agencyID': self.getAgencyID()
+ , 'url': self.getURL(), 'docID': dochash,
+ "date": edate, "title": "Disclosure Log Updated", "description": description}
+ foidocsdb.save(doc)
+ else:
+ print "already saved"
+
+
+class GenericDOCXDisclogScraper(GenericDisclogScraper):
+
+ def doScrape(self):
+ foidocsdb = scrape.couch['disclosr-foidocuments']
+ (url, mime_type, content) = scrape.fetchURL(scrape.docsdb
+ , self.getURL(), "foidocuments", self.getAgencyID())
+ mydoc = zipfile.ZipFile(file)
+ xmlcontent = mydoc.read('word/document.xml')
+ document = etree.fromstring(xmlcontent)
+ ## Fetch all the text out of the document we just created
+ paratextlist = getdocumenttext(document)
+ # Make explicit unicode version
+ newparatextlist = []
+ for paratext in paratextlist:
+ newparatextlist.append(paratext.encode("utf-8"))
+ ## Print our documnts test with two newlines under each paragraph
+ description = '\n\n'.join(newparatextlist).strip(' \t\n\r')
+ dochash = scrape.mkhash(description)
+ doc = foidocsdb.get(dochash)
+
+ if doc is None:
+ print "saving " + dochash
+ edate = time().strftime("%Y-%m-%d")
+ doc = {'_id': dochash, 'agencyID': self.getAgencyID()
+ , 'url': self.getURL(), 'docID': dochash,
+ "date": edate, "title": "Disclosure Log Updated", "description": description}
+ foidocsdb.save(doc)
+ else:
+ print "already saved"
class GenericRSSDisclogScraper(GenericDisclogScraper):
- def doScrape(self):
- foidocsdb = scrape.couch['disclosr-foidocuments']
- (url,mime_type,content) = scrape.fetchURL(scrape.docsdb, self.getURL(), "foidocuments", self.getAgencyID())
- feed = feedparser.parse(content)
- for entry in feed.entries:
- #print entry
- print entry.id
- hash = scrape.mkhash(entry.id)
- #print hash
- doc = foidocsdb.get(hash)
- #print doc
- if doc == None:
- print "saving "+ hash
- edate = datetime.fromtimestamp(mktime( entry.published_parsed)).strftime("%Y-%m-%d")
- doc = {'_id': hash, 'agencyID': self.getAgencyID(), 'url': entry.link, 'docID': entry.id,
- "date": edate,"title": entry.title}
- self.getDescription(entry,entry, doc)
- foidocsdb.save(doc)
+ def doScrape(self):
+ foidocsdb = scrape.couch['disclosr-foidocuments']
+ (url, mime_type, content) = scrape.fetchURL(scrape.docsdb,
+ self.getURL(), "foidocuments", self.getAgencyID())
+ feed = feedparser.parse(content)
+ for entry in feed.entries:
+ #print entry
+ print entry.id
+ dochash = scrape.mkhash(entry.id)
+ doc = foidocsdb.get(dochash)
+ #print doc
+ if doc is None:
+ print "saving " + dochash
+ edate = datetime.fromtimestamp(
+ mktime(entry.published_parsed)).strftime("%Y-%m-%d")
+ doc = {'_id': dochash, 'agencyID': self.getAgencyID(),
+ 'url': entry.link, 'docID': entry.id,
+ "date": edate, "title": entry.title}
+ self.getDescription(entry, entry, doc)
+ foidocsdb.save(doc)
+ else:
+ print "already saved"
+
+ def getDescription(self, content, entry, doc):
+ """ get description from rss entry"""
+ doc.update({'description': content.summary})
+ return
+
+
+class GenericOAICDisclogScraper(GenericDisclogScraper):
+ __metaclass__ = abc.ABCMeta
+
+ @abc.abstractmethod
+ def getColumns(self, columns):
+ """ rearranges columns if required """
+ return
+
+ def getColumnCount(self):
+ return 5
+
+ def getDescription(self, content, entry, doc):
+ """ get description from rss entry"""
+ descriptiontxt = ""
+ for string in content.stripped_strings:
+ descriptiontxt = descriptiontxt + " \n" + string
+ doc.update({'description': descriptiontxt})
+
+ def getTitle(self, content, entry, doc):
+ doc.update({'title': (''.join(content.stripped_strings))})
+
+ def getTable(self, soup):
+ return soup.table
+
+ def getRows(self, table):
+ return table.find_all('tr')
+
+ def getDate(self, content, entry, doc):
+ 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
+
+ def getLinks(self, content, entry, doc):
+ links = []
+ for atag in entry.find_all("a"):
+ if atag.has_key('href'):
+ links.append(scrape.fullurl(content, atag['href']))
+ if links != []:
+ doc.update({'links': links})
+ return
+
+ def doScrape(self):
+ foidocsdb = scrape.couch['disclosr-foidocuments']
+ (url, mime_type, content) = scrape.fetchURL(scrape.docsdb,
+ self.getURL(), "foidocuments", self.getAgencyID())
+ if content is not None:
+ if mime_type is "text/html"\
+ or mime_type is "application/xhtml+xml"\
+ or mime_type is"application/xml":
+ # http://www.crummy.com/software/BeautifulSoup/documentation.html
+ soup = BeautifulSoup(content)
+ table = self.getTable(soup)
+ for row in self.getRows(table):
+ columns = row.find_all('td')
+ if len(columns) is self.getColumnCount():
+ (id, date, title,
+ description, notes) = self.getColumns(columns)
+ print self.remove_control_chars(
+ ''.join(id.stripped_strings))
+ if id.string is None:
+ dochash = scrape.mkhash(
+ self.remove_control_chars(
+ url + (''.join(date.stripped_strings))))
else:
- print "already saved"
- def getDescription(self, content, entry, doc):
- """ get description from rss entry"""
- doc.update({'description': content.summary})
- return
-
-class GenericOAICDisclogScraper(GenericDisclogScraper):
- __metaclass__ = abc.ABCMeta
- @abc.abstractmethod
- def getColumns(self,columns):
- """ rearranges columns if required """
- return
- def getColumnCount(self):
- return 5
- def getDescription(self, content, entry, doc):
- """ get description from rss entry"""
- descriptiontxt = ""
- for string in content.stripped_strings:
- descriptiontxt = descriptiontxt + " \n" + string
- doc.update({'description': descriptiontxt})
- return
- def getTitle(self, content, entry, doc):
- 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):
- 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
- def getLinks(self, content, entry, doc):
- links = []
- for atag in entry.find_all("a"):
- if atag.has_key('href'):
- links.append(scrape.fullurl(content,atag['href']))
- if links != []:
- doc.update({'links': links})
- return
-
- def doScrape(self):
- foidocsdb = scrape.couch['disclosr-foidocuments']
- (url,mime_type,content) = scrape.fetchURL(scrape.docsdb, self.getURL(), "foidocuments", self.getAgencyID())
- if content != 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)
- table = self.getTable(soup)
- for row in self.getRows(table):
- columns = row.find_all('td')
- if len(columns) == self.getColumnCount():
- (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:
- hash = scrape.mkhash(self.remove_control_chars(url+(''.join(id.stripped_strings))))
- doc = foidocsdb.get(hash)
-
- if doc == None:
- print "saving " +hash
- 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': (''.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',
+ dochash = scrape.mkhash(
+ self.remove_control_chars(
+ url + (''.join(id.stripped_strings))))
+ doc = foidocsdb.get(hash)
+
+ if doc is None:
+ print "saving " + hash
+ 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 is not None:
+ 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'] != '':
+ if doc['title'] not in badtitles\
+ and doc['description'] != '':
print "saving"
foidocsdb.save(doc)
- else:
- print "already saved "+hash
-
- elif len(row.find_all('th')) == self.getColumnCount():
- print "header row"
-
- else:
- print "ERROR number of columns incorrect"
- print row
-
+ else:
+ print "already saved " + dochash
+
+ elif len(row.find_all('th')) is self.getColumnCount():
+ print "header row"
+
+ else:
+ print "ERROR number of columns incorrect"
+ print row
+
--- a/documents/rss.xml.php
+++ b/documents/rss.xml.php
@@ -9,11 +9,12 @@
$TestFeed = new RSS2FeedWriter();
//Setting the channel elements
//Use wrapper functions for common channelelements
-$TestFeed->setTitle('Last Modified - All');
+$TestFeed->setTitle('disclosurelo.gs Newest Entries - All');
$TestFeed->setLink('http://disclosurelo.gs/rss.xml.php');
-$TestFeed->setDescription('Latest entries');
- $TestFeed->setChannelElement('language', 'en-us');
- $TestFeed->setChannelElement('pubDate', date(DATE_RSS, time()));
+$TestFeed->setDescription('disclosurelo.gs Newest Entries - All Agencies');
+$TestFeed->setChannelElement('language', 'en-us');
+$TestFeed->setChannelElement('pubDate', date(DATE_RSS, time()));
+
//Retriving informations from database
$idtoname = Array();
$agenciesdb = $server->get_db('disclosr-agencies');
@@ -21,17 +22,18 @@
$idtoname[$row->id] = trim($row->value->name);
}
$foidocsdb = $server->get_db('disclosr-foidocuments');
-$rows = $foidocsdb->get_view("app", "byDate", Array('9999-99-99','0000-00-00', 50), true)->rows;
+$rows = $foidocsdb->get_view("app", "byDate", Array('9999-99-99', '0000-00-00', 50), true)->rows;
//print_r($rows);
foreach ($rows as $row) {
//Create an empty FeedItem
$newItem = $TestFeed->createNewItem();
//Add elements to the feed item
$newItem->setTitle($row->value->title);
- $newItem->setLink("view.php?id=".$row->value->_id);
- $newItem->setDate(date("c", strtotime($row->value->date)));
- $newItem->setDescription(displayLogEntry($row,$idtoname));
- $newItem->addElement('guid', $row->value->_id,array('isPermaLink'=>'true'));
+ $newItem->setLink("http://disclosurelo.gs/view.php?id=" . $row->value->_id);
+ $newItem->setDate(strtotime($row->value->date));
+ $newItem->setDescription(displayLogEntry($row, $idtoname));
+ $newItem->setAuthor($idtoname[$row->value->agencyID]);
+ $newItem->addElement('guid', "http://disclosurelo.gs/view.php?id=" . $row->value->_id, array('isPermaLink' => 'true'));
//Now add the feed item
$TestFeed->addItem($newItem);
}
--- a/documents/scrape.py
+++ b/documents/scrape.py
@@ -8,186 +8,188 @@
import time
import os
import mimetypes
-import re
import urllib
import urlparse
def mkhash(input):
- return hashlib.md5(input).hexdigest().encode("utf-8")
+ return hashlib.md5(input).hexdigest().encode("utf-8")
def canonurl(url):
- r"""Return the canonical, ASCII-encoded form of a UTF-8 encoded URL, or ''
- if the URL looks invalid.
- >>> canonurl('\xe2\x9e\xa1.ws') # tinyarro.ws
- 'http://xn--hgi.ws/'
- """
- # strip spaces at the ends and ensure it's prefixed with 'scheme://'
- url = url.strip()
- if not url:
- return ''
- if not urlparse.urlsplit(url).scheme:
- url = 'http://' + url
-
- # turn it into Unicode
- #try:
- # url = unicode(url, 'utf-8')
- #except UnicodeDecodeError:
- # return '' # bad UTF-8 chars in URL
-
- # parse the URL into its components
- parsed = urlparse.urlsplit(url)
- scheme, netloc, path, query, fragment = parsed
-
- # ensure scheme is a letter followed by letters, digits, and '+-.' chars
- if not re.match(r'[a-z][-+.a-z0-9]*$', scheme, flags=re.I):
- return ''
- scheme = str(scheme)
-
- # ensure domain and port are valid, eg: sub.domain.<1-to-6-TLD-chars>[:port]
- match = re.match(r'(.+\.[a-z0-9]{1,6})(:\d{1,5})?$', netloc, flags=re.I)
- if not match:
- return ''
- domain, port = match.groups()
- netloc = domain + (port if port else '')
- netloc = netloc.encode('idna')
-
- # ensure path is valid and convert Unicode chars to %-encoded
- if not path:
- path = '/' # eg: 'http://google.com' -> 'http://google.com/'
- path = urllib.quote(urllib.unquote(path.encode('utf-8')), safe='/;')
-
- # ensure query is valid
- query = urllib.quote(urllib.unquote(query.encode('utf-8')), safe='=&?/')
-
- # ensure fragment is valid
- fragment = urllib.quote(urllib.unquote(fragment.encode('utf-8')))
-
- # piece it all back together, truncating it to a maximum of 4KB
- url = urlparse.urlunsplit((scheme, netloc, path, query, fragment))
- return url[:4096]
+ r"""Return the canonical, ASCII-encoded form of a UTF-8 encoded URL, or ''
+ if the URL looks invalid.
+ >>> canonurl('\xe2\x9e\xa1.ws') # tinyarro.ws
+ 'http://xn--hgi.ws/'
+ """
+ # strip spaces at the ends and ensure it's prefixed with 'scheme://'
+ url = url.strip()
+ if not url:
+ return ''
+ if not urlparse.urlsplit(url).scheme:
+ url = 'http://' + url
+
+ # turn it into Unicode
+ #try:
+ # url = unicode(url, 'utf-8')
+ #except UnicodeDecodeError:
+ # return '' # bad UTF-8 chars in URL
+
+ # parse the URL into its components
+ parsed = urlparse.urlsplit(url)
+ scheme, netloc, path, query, fragment = parsed
+
+ # ensure scheme is a letter followed by letters, digits, and '+-.' chars
+ if not re.match(r'[a-z][-+.a-z0-9]*$', scheme, flags=re.I):
+ return ''
+ scheme = str(scheme)
+
+ # ensure domain and port are valid, eg: sub.domain.<1-to-6-TLD-chars>[:port]
+ match = re.match(r'(.+\.[a-z0-9]{1,6})(:\d{1,5})?$', netloc, flags=re.I)
+ if not match:
+ return ''
+ domain, port = match.groups()
+ netloc = domain + (port if port else '')
+ netloc = netloc.encode('idna')
+
+ # ensure path is valid and convert Unicode chars to %-encoded
+ if not path:
+ path = '/' # eg: 'http://google.com' -> 'http://google.com/'
+ path = urllib.quote(urllib.unquote(path.encode('utf-8')), safe='/;')
+
+ # ensure query is valid
+ query = urllib.quote(urllib.unquote(query.encode('utf-8')), safe='=&?/')
+
+ # ensure fragment is valid
+ fragment = urllib.quote(urllib.unquote(fragment.encode('utf-8')))
+
+ # piece it all back together, truncating it to a maximum of 4KB
+ url = urlparse.urlunsplit((scheme, netloc, path, query, fragment))
+ return url[:4096]
def fullurl(url,href):
- href = href.replace(" ","%20")
- href = re.sub('#.*$','',href)
- return urljoin(url,href)
+ href = href.replace(" ","%20")
+ href = re.sub('#.*$','',href)
+ return urljoin(url,href)
#http://diveintopython.org/http_web_services/etags.html
-class NotModifiedHandler(urllib2.BaseHandler):
- def http_error_304(self, req, fp, code, message, headers):
- addinfourl = urllib2.addinfourl(fp, headers, req.get_full_url())
- addinfourl.code = code
- return addinfourl
+class NotModifiedHandler(urllib2.BaseHandler):
+ def http_error_304(self, req, fp, code, message, headers):
+ addinfourl = urllib2.addinfourl(fp, headers, req.get_full_url())
+ addinfourl.code = code
+ return addinfourl
def fetchURL(docsdb, url, fieldName, agencyID, scrape_again=True):
- url = canonurl(url)
- hash = mkhash(url)
- req = urllib2.Request(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)
- doc = docsdb.get(hash)
- if doc == None:
- doc = {'_id': hash, 'agencyID': agencyID, 'url': url, 'fieldName':fieldName}
- else:
- if (('page_scraped' in doc) and (time.time() - doc['page_scraped']) < 60*24*14*1000):
- 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)
- 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)
-
- time.sleep(3) # wait 3 seconds to give webserver time to recover
-
- req.add_header("User-Agent", "Mozilla/4.0 (compatible; Prometheus webspider; owner maxious@lambdacomplex.org)")
- #if there is a previous version stored in couchdb, load caching helper tags
- if doc.has_key('etag'):
- req.add_header("If-None-Match", doc['etag'])
- if doc.has_key('last_modified'):
- req.add_header("If-Modified-Since", doc['last_modified'])
-
- opener = urllib2.build_opener(NotModifiedHandler())
- try:
- url_handle = opener.open(req)
- doc['url'] = url_handle.geturl() # may have followed a redirect to a new url
- headers = url_handle.info() # the addinfourls have the .info() too
- doc['etag'] = headers.getheader("ETag")
- doc['last_modified'] = headers.getheader("Last-Modified")
- doc['date'] = headers.getheader("Date")
- doc['page_scraped'] = time.time()
- doc['web_server'] = headers.getheader("Server")
- doc['via'] = headers.getheader("Via")
- doc['powered_by'] = headers.getheader("X-Powered-By")
- doc['file_size'] = headers.getheader("Content-Length")
- content_type = headers.getheader("Content-Type")
- if content_type != None:
- doc['mime_type'] = content_type.split(";")[0]
- else:
- (type,encoding) = mimetypes.guess_type(url)
- doc['mime_type'] = type
- if hasattr(url_handle, 'code'):
- if url_handle.code == 304:
- print "the web page has not been modified"
- return (None,None,None)
- else:
- content = url_handle.read()
- docsdb.save(doc)
- doc = docsdb.get(hash) # need to get a _rev
- docsdb.put_attachment(doc, content, str(time.time())+"-"+os.path.basename(url), doc['mime_type'])
- return (doc['url'], doc['mime_type'], content)
- #store as attachment epoch-filename
-
- except urllib2.URLError as e:
- error = ""
- if hasattr(e, 'reason'):
- error = "error %s in downloading %s" % (str(e.reason), url)
- elif hasattr(e, 'code'):
- error = "error %s in downloading %s" % (e.code, url)
- print error
- doc['error'] = error
- docsdb.save(doc)
- return (None,None,None)
+ url = canonurl(url)
+ hash = mkhash(url)
+ req = urllib2.Request(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)
+ doc = docsdb.get(hash)
+ if doc == None:
+ doc = {'_id': hash, 'agencyID': agencyID, 'url': url, 'fieldName':fieldName}
+ else:
+ if (('page_scraped' in doc) and (time.time() - doc['page_scraped']) < 60*24*14*1000):
+ print "Uh oh, trying to scrape URL again too soon!"+hash
+ last_attachment_fname = doc["_attachments"].keys()[-1]
+ last_attachment = docsdb.get_attachment(doc,last_attachment_fname)
+ 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)
+
+ req.add_header("User-Agent", "Mozilla/4.0 (compatible; Prometheus webspider; owner maxious@lambdacomplex.org)")
+ #if there is a previous version stored in couchdb, load caching helper tags
+ if doc.has_key('etag'):
+ req.add_header("If-None-Match", doc['etag'])
+ if doc.has_key('last_modified'):
+ req.add_header("If-Modified-Since", doc['last_modified'])
+
+ opener = urllib2.build_opener(NotModifiedHandler())
+ try:
+ url_handle = opener.open(req)
+ doc['url'] = url_handle.geturl() # may have followed a redirect to a new url
+ headers = url_handle.info() # the addinfourls have the .info() too
+ doc['etag'] = headers.getheader("ETag")
+ doc['last_modified'] = headers.getheader("Last-Modified")
+ doc['date'] = headers.getheader("Date")
+ doc['page_scraped'] = time.time()
+ doc['web_server'] = headers.getheader("Server")
+ doc['via'] = headers.getheader("Via")
+ doc['powered_by'] = headers.getheader("X-Powered-By")
+ doc['file_size'] = headers.getheader("Content-Length")
+ content_type = headers.getheader("Content-Type")
+ if content_type != None:
+ doc['mime_type'] = content_type.split(";")[0]
+ else:
+ (type,encoding) = mimetypes.guess_type(url)
+ doc['mime_type'] = type
+ if hasattr(url_handle, 'code'):
+ if url_handle.code == 304:
+ print "the web page has not been modified"+hash
+ last_attachment_fname = doc["_attachments"].keys()[-1]
+ last_attachment = docsdb.get_attachment(doc,last_attachment_fname)
+ content = last_attachment
+ return (doc['url'],doc['mime_type'],content)
+ else:
+ print "new webpage loaded"
+ content = url_handle.read()
+ docsdb.save(doc)
+ doc = docsdb.get(hash) # need to get a _rev
+ docsdb.put_attachment(doc, content, str(time.time())+"-"+os.path.basename(url), doc['mime_type'])
+ return (doc['url'], doc['mime_type'], content)
+ #store as attachment epoch-filename
+
+ except urllib2.URLError as e:
+ print "error!"
+ error = ""
+ if hasattr(e, 'reason'):
+ error = "error %s in downloading %s" % (str(e.reason), url)
+ elif hasattr(e, 'code'):
+ error = "error %s in downloading %s" % (e.code, url)
+ print error
+ doc['error'] = error
+ docsdb.save(doc)
+ return (None,None,None)
def scrapeAndStore(docsdb, url, depth, fieldName, agencyID):
- (url,mime_type,content) = fetchURL(docsdb, url, fieldName, agencyID)
- badURLs = ["http://www.ausport.gov.au/supporting/funding/grants_and_scholarships/grant_funding_report"]
- if content != None and depth > 0 and url != "http://www.ausport.gov.au/supporting/funding/grants_and_scholarships/grant_funding_report":
- 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)
- navIDs = soup.findAll(id=re.compile('nav|Nav|menu|bar|left|right|sidebar|more-links|breadcrumb|footer|header'))
- for nav in navIDs:
- print "Removing element", nav['id']
- nav.extract()
- navClasses = soup.findAll(attrs={'class' : re.compile('nav|menu|bar|left|right|sidebar|more-links|breadcrumb|footer|header')})
- for nav in navClasses:
- print "Removing element", nav['class']
- nav.extract()
- links = soup.findAll('a') # soup.findAll('a', id=re.compile("^p-"))
- linkurls = set([])
- for link in links:
- if link.has_key("href"):
- if link['href'].startswith("http"):
- # lets not do externa