import sys | import sys |
import os | import os |
sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../')) | sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../')) |
import scrape | import scrape |
from bs4 import BeautifulSoup | from bs4 import BeautifulSoup |
from time import mktime | from time import mktime |
import feedparser | import feedparser |
import abc | import abc |
import unicodedata | import unicodedata |
import re | import re |
import dateutil | import dateutil |
from dateutil.parser import * | from dateutil.parser import * |
from datetime import * | from datetime import * |
import codecs | import codecs |
import difflib | import difflib |
from StringIO import StringIO | from StringIO import StringIO |
from pdfminer.pdfparser import PDFDocument, PDFParser | from pdfminer.pdfparser import PDFDocument, PDFParser |
from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter, process_pdf | from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter, process_pdf |
from pdfminer.pdfdevice import PDFDevice, TagExtractor | from pdfminer.pdfdevice import PDFDevice, TagExtractor |
from pdfminer.converter import TextConverter | from pdfminer.converter import TextConverter |
from pdfminer.cmapdb import CMapDB | from pdfminer.cmapdb import CMapDB |
from pdfminer.layout import LAParams | from pdfminer.layout import LAParams |
class GenericDisclogScraper(object): | class GenericDisclogScraper(object): |
__metaclass__ = abc.ABCMeta | __metaclass__ = abc.ABCMeta |
agencyID = None | agencyID = None |
disclogURL = None | disclogURL = None |
def remove_control_chars(self, input): | def remove_control_chars(self, input): |
return "".join([i for i in input if ord(i) in range(32, 127)]) | return "".join([i for i in input if ord(i) in range(32, 127)]) |
def getAgencyID(self): | def getAgencyID(self): |
""" disclosr agency id """ | """ disclosr agency id """ |
if self.agencyID is None: | if self.agencyID is None: |
self.agencyID = os.path.basename(sys.argv[0]).replace(".py", "") | self.agencyID = os.path.basename(sys.argv[0]).replace(".py", "") |
return self.agencyID | return self.agencyID |
def getURL(self): | def getURL(self): |
""" disclog URL""" | """ disclog URL""" |
if self.disclogURL is None: | if self.disclogURL is None: |
agency = scrape.agencydb.get(self.getAgencyID()) | agency = scrape.agencydb.get(self.getAgencyID()) |
self.disclogURL = agency['FOIDocumentsURL'] | self.disclogURL = agency['FOIDocumentsURL'] |
return self.disclogURL | return self.disclogURL |
@abc.abstractmethod | @abc.abstractmethod |
def doScrape(self): | def doScrape(self): |
""" do the scraping """ | """ do the scraping """ |
return | return |
class GenericHTMLDisclogScraper(GenericDisclogScraper): | class GenericHTMLDisclogScraper(GenericDisclogScraper): |
def doScrape(self): | def doScrape(self): |
foidocsdb = scrape.couch['disclosr-foidocuments'] | foidocsdb = scrape.couch['disclosr-foidocuments'] |
(url, mime_type, rcontent) = scrape.fetchURL(scrape.docsdb, | (url, mime_type, rcontent) = scrape.fetchURL(scrape.docsdb, |
self.getURL(), "foidocuments", self.getAgencyID()) | self.getURL(), "foidocuments", self.getAgencyID()) |
content = rcontent | content = rcontent |
dochash = scrape.mkhash(content) | dochash = scrape.mkhash(content) |
doc = foidocsdb.get(dochash) | doc = foidocsdb.get(dochash) |
if doc is None: | if doc is None: |
print "saving " + dochash | print "saving " + dochash |
description = "This log may have updated but as it was not in a table last time we viewed it, we cannot extract what has changed. Please refer to the agency's website Disclosure Log to see the most recent entries" | description = "This log may have updated but as it was not in a table last time we viewed it, we cannot extract what has changed. Please refer to the agency's website Disclosure Log to see the most recent entries" |
last_attach = scrape.getLastAttachment(scrape.docsdb, self.getURL()) | last_attach = scrape.getLastAttachment(scrape.docsdb, self.getURL()) |
if last_attach != None: | if last_attach != None: |
html_diff = difflib.HtmlDiff() | html_diff = difflib.HtmlDiff() |
diff = html_diff.make_table(last_attach.read().split('\n'), | diff = html_diff.make_table(last_attach.read().split('\n'), |
content.split('\n')) | content.split('\n')) |
edate = date.today().strftime("%Y-%m-%d") | edate = date.today().strftime("%Y-%m-%d") |
doc = {'_id': dochash, 'agencyID': self.getAgencyID() | doc = {'_id': dochash, 'agencyID': self.getAgencyID() |
, 'url': self.getURL(), 'docID': dochash, | , 'url': self.getURL(), 'docID': dochash, |
"date": edate, "title": "Disclosure Log Updated", "description": self.remove_control_chars(description), "diff": diff} | "date": edate, "title": "Disclosure Log Updated", |
"description": self.remove_control_chars(description), "diff": self.remove_control_chars(diff)} | |
foidocsdb.save(doc) | foidocsdb.save(doc) |
else: | else: |
print "already saved" | print "already saved" |
class GenericPDFDisclogScraper(GenericDisclogScraper): | class GenericPDFDisclogScraper(GenericDisclogScraper): |
def doScrape(self): | def doScrape(self): |
foidocsdb = scrape.couch['disclosr-foidocuments'] | foidocsdb = scrape.couch['disclosr-foidocuments'] |
(url, mime_type, content) = scrape.fetchURL(scrape.docsdb, | (url, mime_type, content) = scrape.fetchURL(scrape.docsdb, |
self.getURL(), "foidocuments", self.getAgencyID()) | self.getURL(), "foidocuments", self.getAgencyID()) |
laparams = LAParams() | laparams = LAParams() |
rsrcmgr = PDFResourceManager(caching=True) | rsrcmgr = PDFResourceManager(caching=True) |
outfp = StringIO() | outfp = StringIO() |
device = TextConverter(rsrcmgr, outfp, codec='utf-8', | device = TextConverter(rsrcmgr, outfp, codec='utf-8', |
laparams=laparams) | laparams=laparams) |
fp = StringIO() | fp = StringIO() |
fp.write(content) | fp.write(content) |
process_pdf(rsrcmgr, device, fp, set(), caching=True, | process_pdf(rsrcmgr, device, fp, set(), caching=True, |
check_extractable=True) | check_extractable=True) |
description = outfp.getvalue() | description = outfp.getvalue() |
fp.close() | fp.close() |
device.close() | device.close() |
outfp.close() | outfp.close() |
dochash = scrape.mkhash(description) | dochash = scrape.mkhash(description) |
doc = foidocsdb.get(dochash) | doc = foidocsdb.get(dochash) |
if doc is None: | if doc is None: |
print "saving " + dochash | print "saving " + dochash |
edate = date.today().strftime("%Y-%m-%d") | edate = date.today().strftime("%Y-%m-%d") |
doc = {'_id': dochash, 'agencyID': self.getAgencyID() | doc = {'_id': dochash, 'agencyID': self.getAgencyID() |
, 'url': self.getURL(), 'docID': dochash, | , 'url': self.getURL(), 'docID': dochash, |
"date": edate, "title": "Disclosure Log Updated", "description": self.remove_control_chars(description)} | "date": edate, "title": "Disclosure Log Updated", "description": self.remove_control_chars(description)} |
foidocsdb.save(doc) | foidocsdb.save(doc) |
else: | else: |
print "already saved" | print "already saved" |
class GenericDOCXDisclogScraper(GenericDisclogScraper): | class GenericDOCXDisclogScraper(GenericDisclogScraper): |
def doScrape(self): | def doScrape(self): |
foidocsdb = scrape.couch['disclosr-foidocuments'] | foidocsdb = scrape.couch['disclosr-foidocuments'] |
(url, mime_type, content) = scrape.fetchURL(scrape.docsdb | (url, mime_type, content) = scrape.fetchURL(scrape.docsdb |
, self.getURL(), "foidocuments", self.getAgencyID()) | , self.getURL(), "foidocuments", self.getAgencyID()) |
mydoc = zipfile.ZipFile(file) | mydoc = zipfile.ZipFile(file) |
xmlcontent = mydoc.read('word/document.xml') | xmlcontent = mydoc.read('word/document.xml') |
document = etree.fromstring(xmlcontent) | document = etree.fromstring(xmlcontent) |
## Fetch all the text out of the document we just created | ## Fetch all the text out of the document we just created |
paratextlist = getdocumenttext(document) | paratextlist = getdocumenttext(document) |
# Make explicit unicode version | # Make explicit unicode version |
newparatextlist = [] | newparatextlist = [] |
for paratext in paratextlist: | for paratext in paratextlist: |
newparatextlist.append(paratext.encode("utf-8")) | newparatextlist.append(paratext.encode("utf-8")) |
## Print our documnts test with two newlines under each paragraph | ## Print our documnts test with two newlines under each paragraph |
description = '\n\n'.join(newparatextlist).strip(' \t\n\r') | description = '\n\n'.join(newparatextlist).strip(' \t\n\r') |
dochash = scrape.mkhash(description) | dochash = scrape.mkhash(description) |
doc = foidocsdb.get(dochash) | doc = foidocsdb.get(dochash) |
if doc is None: | if doc is None: |
print "saving " + dochash | print "saving " + dochash |
edate = time().strftime("%Y-%m-%d") | edate = time().strftime("%Y-%m-%d") |
doc = {'_id': dochash, 'agencyID': self.getAgencyID() | doc = {'_id': dochash, 'agencyID': self.getAgencyID() |
, 'url': self.getURL(), 'docID': dochash, | , 'url': self.getURL(), 'docID': dochash, |
"date": edate, "title": "Disclosure Log Updated", "description": description} | "date": edate, "title": "Disclosure Log Updated", "description": description} |
foidocsdb.save(doc) | foidocsdb.save(doc) |
else: | else: |
print "already saved" | print "already saved" |
class GenericRSSDisclogScraper(GenericDisclogScraper): | class GenericRSSDisclogScraper(GenericDisclogScraper): |
def doScrape(self): | def doScrape(self): |
foidocsdb = scrape.couch['disclosr-foidocuments'] | foidocsdb = scrape.couch['disclosr-foidocuments'] |
(url, mime_type, content) = scrape.fetchURL(scrape.docsdb, | (url, mime_type, content) = scrape.fetchURL(scrape.docsdb, |
self.getURL(), "foidocuments", self.getAgencyID()) | self.getURL(), "foidocuments", self.getAgencyID()) |
feed = feedparser.parse(content) | feed = feedparser.parse(content) |
for entry in feed.entries: | for entry in feed.entries: |
#print entry | #print entry |
print entry.id | print entry.id |
dochash = scrape.mkhash(entry.id) | dochash = scrape.mkhash(entry.id) |
doc = foidocsdb.get(dochash) | doc = foidocsdb.get(dochash) |
#print doc | #print doc |
if doc is None: | if doc is None: |
print "saving " + dochash | print "saving " + dochash |
edate = datetime.fromtimestamp( | edate = datetime.fromtimestamp( |
mktime(entry.published_parsed)).strftime("%Y-%m-%d") | mktime(entry.published_parsed)).strftime("%Y-%m-%d") |
doc = {'_id': dochash, 'agencyID': self.getAgencyID(), | doc = {'_id': dochash, 'agencyID': self.getAgencyID(), |
'url': entry.link, 'docID': entry.id, | 'url': entry.link, 'docID': entry.id, |
"date": edate, "title": entry.title} | "date": edate, "title": entry.title} |
self.getDescription(entry, entry, doc) | self.getDescription(entry, entry, doc) |
foidocsdb.save(doc) | foidocsdb.save(doc) |
else: | else: |
print "already saved" | print "already saved" |
def getDescription(self, content, entry, doc): | def getDescription(self, content, entry, doc): |
""" get description from rss entry""" | """ get description from rss entry""" |
doc.update({'description': content.summary}) | doc.update({'description': content.summary}) |
return | return |
class GenericOAICDisclogScraper(GenericDisclogScraper): | class GenericOAICDisclogScraper(GenericDisclogScraper): |
__metaclass__ = abc.ABCMeta | __metaclass__ = abc.ABCMeta |
@abc.abstractmethod | @abc.abstractmethod |
def getColumns(self, columns): | def getColumns(self, columns): |
""" rearranges columns if required """ | """ rearranges columns if required """ |
return | return |
def getColumnCount(self): | def getColumnCount(self): |
return 5 | return 5 |
def getDescription(self, content, entry, doc): | def getDescription(self, content, entry, doc): |
""" get description from rss entry""" | """ get description from rss entry""" |
descriptiontxt = "" | descriptiontxt = "" |
for string in content.stripped_strings: | for string in content.stripped_strings: |
descriptiontxt = descriptiontxt + " \n" + string | descriptiontxt = descriptiontxt + " \n" + string |
doc.update({'description': descriptiontxt}) | doc.update({'description': descriptiontxt}) |
def getTitle(self, content, entry, doc): | def getTitle(self, content, entry, doc): |
doc.update({'title': (''.join(content.stripped_strings))}) | doc.update({'title': (''.join(content.stripped_strings))}) |
def getTable(self, soup): | def getTable(self, soup): |
return soup.table | return soup.table |
def getRows(self, table): | def getRows(self, table): |
return table.find_all('tr') | return table.find_all('tr') |
def getDate(self, content, entry, doc): | def getDate(self, content, entry, doc): |
date = ''.join(content.stripped_strings).strip() | date = ''.join(content.stripped_strings).strip() |
(a, b, c) = date.partition("(") | (a, b, c) = date.partition("(") |
date = self.remove_control_chars(a.replace("Octber", "October").replace("1012","2012")) | date = self.remove_control_chars(a.replace("Octber", "October").replace("1012","2012")) |
print date | print date |
edate = parse(date, dayfirst=True, fuzzy=True).strftime("%Y-%m-%d") | edate = parse(date, dayfirst=True, fuzzy=True).strftime("%Y-%m-%d") |
print edate | print edate |
doc.update({'date': edate}) | doc.update({'date': edate}) |
return | return |
def getLinks(self, content, entry, doc): | def getLinks(self, content, entry, doc): |
links = [] | links = [] |
for atag in entry.find_all("a"): | for atag in entry.find_all("a"): |
if atag.has_key('href'): | if atag.has_key('href'): |
links.append(scrape.fullurl(content, atag['href'])) | links.append(scrape.fullurl(content, atag['href'])) |
if links != []: | if links != []: |
doc.update({'links': links}) | doc.update({'links': links}) |
return | return |
def doScrape(self): | def doScrape(self): |
foidocsdb = scrape.couch['disclosr-foidocuments'] | foidocsdb = scrape.couch['disclosr-foidocuments'] |
(url, mime_type, content) = scrape.fetchURL(scrape.docsdb, | (url, mime_type, content) = scrape.fetchURL(scrape.docsdb, |
self.getURL(), "foidocuments", self.getAgencyID()) | self.getURL(), "foidocuments", self.getAgencyID()) |
if content is not None: | if content is not None: |
if mime_type == "text/html" or mime_type == "application/xhtml+xml" or mime_type == "application/xml": | if mime_type == "text/html" or mime_type == "application/xhtml+xml" or mime_type == "application/xml": |
# http://www.crummy.com/software/BeautifulSoup/documentation.html | # http://www.crummy.com/software/BeautifulSoup/documentation.html |
print "parsing" | print "parsing" |
soup = BeautifulSoup(content) | soup = BeautifulSoup(content) |
table = self.getTable(soup) | table = self.getTable(soup) |
for row in self.getRows(table): | for row in self.getRows(table): |
columns = row.find_all('td') | columns = row.find_all('td') |
if len(columns) is self.getColumnCount(): | if len(columns) is self.getColumnCount(): |
(id, date, title, | (id, date, title, |
description, notes) = self.getColumns(columns) | description, notes) = self.getColumns(columns) |
print self.remove_control_chars( | print self.remove_control_chars( |
''.join(id.stripped_strings)) | ''.join(id.stripped_strings)) |
if id.string is None: | if id.string is None: |
dochash = scrape.mkhash( | dochash = scrape.mkhash( |
self.remove_control_chars( | self.remove_control_chars( |
url + (''.join(date.stripped_strings)))) | url + (''.join(date.stripped_strings)))) |
else: | else: |
dochash = scrape.mkhash( | dochash = scrape.mkhash( |
self.remove_control_chars( | self.remove_control_chars( |
url + (''.join(id.stripped_strings)))) | url + (''.join(id.stripped_strings)))) |
doc = foidocsdb.get(dochash) | doc = foidocsdb.get(dochash) |
if doc is None: | if doc is None: |
print "saving " + dochash | print "saving " + dochash |
doc = {'_id': dochash, | doc = {'_id': dochash, |
'agencyID': self.getAgencyID(), | 'agencyID': self.getAgencyID(), |
'url': self.getURL(), | 'url': self.getURL(), |
'docID': (''.join(id.stripped_strings))} | 'docID': (''.join(id.stripped_strings))} |
self.getLinks(self.getURL(), row, doc) | self.getLinks(self.getURL(), row, doc) |
self.getTitle(title, row, doc) | self.getTitle(title, row, doc) |
self.getDate(date, row, doc) | self.getDate(date, row, doc) |
self.getDescription(description, row, doc) | self.getDescription(description, row, doc) |
if notes is not None: | if notes is not None: |
doc.update({'notes': ( | doc.update({'notes': ( |
''.join(notes.stripped_strings))}) | ''.join(notes.stripped_strings))}) |
badtitles = ['-', 'Summary of FOI Request' | badtitles = ['-', 'Summary of FOI Request' |
, 'FOI request(in summary form)' | , 'FOI request(in summary form)' |
, 'Summary of FOI request received by the ASC', | , 'Summary of FOI request received by the ASC', |
'Summary of FOI request received by agency/minister', | 'Summary of FOI request received by agency/minister', |
'Description of Documents Requested', 'FOI request', | 'Description of Documents Requested', 'FOI request', |
'Description of FOI Request', 'Summary of request', 'Description', 'Summary', | 'Description of FOI Request', 'Summary of request', 'Description', 'Summary', |
'Summary of FOIrequest received by agency/minister', | 'Summary of FOIrequest received by agency/minister', |
'Summary of FOI request received', 'Description of FOI Request', | 'Summary of FOI request received', 'Description of FOI Request', |
"FOI request", 'Results 1 to 67 of 67'] | "FOI request", 'Results 1 to 67 of 67'] |
if doc['title'] not in badtitles\ | if doc['title'] not in badtitles\ |
and doc['description'] != '': | and doc['description'] != '': |
print "saving" | print "saving" |
foidocsdb.save(doc) | foidocsdb.save(doc) |
else: | else: |
print "already saved " + dochash | print "already saved " + dochash |
elif len(row.find_all('th')) is self.getColumnCount(): | elif len(row.find_all('th')) is self.getColumnCount(): |
print "header row" | print "header row" |
else: | else: |
print "ERROR number of columns incorrect" | print >> sys.stderr, "ERROR number of columns incorrect" |
print row | print row |
for f in scrapers/*.py; | echo "" > /tmp/disclosr-error |
do echo "Processing $f file.."; | for f in scrapers/*.py; do |
python $f; | echo "Processing $f file.."; |
md5=`md5sum /tmp/disclosr-error` | |
python $f 3>&1 1>&2 2>&3 | tee --append /tmp/disclosr-error; | |
md52=`md5sum /tmp/disclosr-error` | |
if [ "$md5" != "$md52" ]; then | |
echo "^^^^^^^^^^^^^^ $f" >> /tmp/disclosr-error; | |
fi | |
if [ "$?" -ne "0" ]; then | if [ "$?" -ne "0" ]; then |
echo "error"; | echo "error"; |
sleep 2; | sleep 1; |
fi | fi |
done | done |
if [ -s /tmp/disclosr-error ] ; then | |
echo "emailling logs.."; | |
mail -E -s "Disclosr errors" maxious@lambdacomplex.org < /tmp/disclosr-error ; | |
fi | |
#http://packages.python.org/CouchDB/client.html | #http://packages.python.org/CouchDB/client.html |
import couchdb | import couchdb |
import urllib2 | import urllib2 |
from BeautifulSoup import BeautifulSoup | from BeautifulSoup import BeautifulSoup |
import re | import re |
import hashlib | import hashlib |
from urlparse import urljoin | from urlparse import urljoin |
import time | import time |
import os | import os |
import sys | |
import mimetypes | import mimetypes |
import urllib | import urllib |
import urlparse | import urlparse |
import socket | import socket |
#couch = couchdb.Server('http://192.168.1.148:5984/') | #couch = couchdb.Server('http://192.168.1.148:5984/') |
couch = couchdb.Server('http://192.168.1.113:5984/') | #couch = couchdb.Server('http://192.168.1.113:5984/') |
#couch = couchdb.Server('http://127.0.0.1:5984/') | couch = couchdb.Server('http://127.0.0.1:5984/') |
def mkhash(input): | def mkhash(input): |
return hashlib.md5(input).hexdigest().encode("utf-8") | return hashlib.md5(input).hexdigest().encode("utf-8") |
def canonurl(url): | def canonurl(url): |
r"""Return the canonical, ASCII-encoded form of a UTF-8 encoded URL, or '' | r"""Return the canonical, ASCII-encoded form of a UTF-8 encoded URL, or '' |
if the URL looks invalid. | if the URL looks invalid. |
>>> canonurl('\xe2\x9e\xa1.ws') # tinyarro.ws | >>> canonurl('\xe2\x9e\xa1.ws') # tinyarro.ws |
'http://xn--hgi.ws/' | 'http://xn--hgi.ws/' |
""" | """ |
# strip spaces at the ends and ensure it's prefixed with 'scheme://' | # strip spaces at the ends and ensure it's prefixed with 'scheme://' |
url = url.strip() | url = url.strip() |
if not url: | if not url: |
return '' | return '' |
if not urlparse.urlsplit(url).scheme: | if not urlparse.urlsplit(url).scheme: |
url = 'http://' + url | url = 'http://' + url |
# turn it into Unicode | # turn it into Unicode |
#try: | #try: |
# url = unicode(url, 'utf-8') | # url = unicode(url, 'utf-8') |
#except UnicodeDecodeError: | #except UnicodeDecodeError: |
# return '' # bad UTF-8 chars in URL | # return '' # bad UTF-8 chars in URL |
# parse the URL into its components | # parse the URL into its components |
parsed = urlparse.urlsplit(url) | parsed = urlparse.urlsplit(url) |
scheme, netloc, path, query, fragment = parsed | scheme, netloc, path, query, fragment = parsed |
# ensure scheme is a letter followed by letters, digits, and '+-.' chars | # 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): | if not re.match(r'[a-z][-+.a-z0-9]*$', scheme, flags=re.I): |
return '' | return '' |
scheme = str(scheme) | scheme = str(scheme) |
# ensure domain and port are valid, eg: sub.domain.<1-to-6-TLD-chars>[:port] | # ensure domain and port are valid, eg: sub.domain.<1-to-6-TLD-chars>[:port] |
match = re.match(r'(.+\.[a-z0-9]{1,6})(:\d{1,5})?$', netloc, flags=re.I) | match = re.match(r'(.+\.[a-z0-9]{1,6})(:\d{1,5})?$', netloc, flags=re.I) |
if not match: | if not match: |
return '' | return '' |
domain, port = match.groups() | domain, port = match.groups() |
netloc = domain + (port if port else '') | netloc = domain + (port if port else '') |
netloc = netloc.encode('idna') | netloc = netloc.encode('idna') |
# ensure path is valid and convert Unicode chars to %-encoded | # ensure path is valid and convert Unicode chars to %-encoded |
if not path: | if not path: |
path = '/' # eg: 'http://google.com' -> 'http://google.com/' | path = '/' # eg: 'http://google.com' -> 'http://google.com/' |
path = urllib.quote(urllib.unquote(path.encode('utf-8')), safe='/;') | path = urllib.quote(urllib.unquote(path.encode('utf-8')), safe='/;') |
# ensure query is valid | # ensure query is valid |
query = urllib.quote(urllib.unquote(query.encode('utf-8')), safe='=&?/') | query = urllib.quote(urllib.unquote(query.encode('utf-8')), safe='=&?/') |
# ensure fragment is valid | # ensure fragment is valid |
fragment = urllib.quote(urllib.unquote(fragment.encode('utf-8'))) | fragment = urllib.quote(urllib.unquote(fragment.encode('utf-8'))) |
# piece it all back together, truncating it to a maximum of 4KB | # piece it all back together, truncating it to a maximum of 4KB |
url = urlparse.urlunsplit((scheme, netloc, path, query, fragment)) | url = urlparse.urlunsplit((scheme, netloc, path, query, fragment)) |
return url[:4096] | return url[:4096] |
def fullurl(url, href): | def fullurl(url, href): |
href = href.replace(" ", "%20") | href = href.replace(" ", "%20") |
href = re.sub('#.*$', '', href) | href = re.sub('#.*$', '', href) |
return urljoin(url, href) | return urljoin(url, href) |
#http://diveintopython.org/http_web_services/etags.html | #http://diveintopython.org/http_web_services/etags.html |
class NotModifiedHandler(urllib2.BaseHandler): | class NotModifiedHandler(urllib2.BaseHandler): |
def http_error_304(self, req, fp, code, message, headers): | def http_error_304(self, req, fp, code, message, headers): |
addinfourl = urllib2.addinfourl(fp, headers, req.get_full_url()) | addinfourl = urllib2.addinfourl(fp, headers, req.get_full_url()) |
addinfourl.code = code | addinfourl.code = code |
return addinfourl | return addinfourl |
def getLastAttachment(docsdb, url): | def getLastAttachment(docsdb, url): |
hash = mkhash(url) | hash = mkhash(url) |
doc = docsdb.get(hash) | doc = docsdb.get(hash) |
if doc != None: | if doc != None: |
last_attachment_fname = doc["_attachments"].keys()[-1] | last_attachment_fname = doc["_attachments"].keys()[-1] |
last_attachment = docsdb.get_attachment(doc, last_attachment_fname) | last_attachment = docsdb.get_attachment(doc, last_attachment_fname) |
return last_attachment | return last_attachment |
else: | else: |
return None | return None |
def fetchURL(docsdb, url, fieldName, agencyID, scrape_again=True): | def fetchURL(docsdb, url, fieldName, agencyID, scrape_again=True): |
url = canonurl(url) | url = canonurl(url) |
hash = mkhash(url) | hash = mkhash(url) |
req = urllib2.Request(url) | req = urllib2.Request(url) |
print "Fetching %s (%s)" % (url, hash) | print "Fetching %s (%s)" % (url, hash) |
if url.startswith("mailto") or url.startswith("javascript") or url.startswith("#") or url == None or url == "": | if url.startswith("mailto") or url.startswith("javascript") or url.startswith("#") or url == None or url == "": |
print "Not a valid HTTP url" | print >> sys.stderr, "Not a valid HTTP url" |
return (None, None, None) | return (None, None, None) |
doc = docsdb.get(hash) | doc = docsdb.get(hash) |
if doc == None: | if doc == None: |
doc = {'_id': hash, 'agencyID': agencyID, 'url': url, 'fieldName': fieldName, 'type': 'website'} | doc = {'_id': hash, 'agencyID': agencyID, 'url': url, 'fieldName': fieldName, 'type': 'website'} |
else: | else: |
if (('page_scraped' in doc) and ((time.time() - doc['page_scraped']) < 60 * 24 * 14) or (scrape_again == False)): | if (('page_scraped' in doc) and ((time.time() - doc['page_scraped']) < 60 * 24 * 14) or (scrape_again == False)): |
print "Uh oh, trying to scrape URL again too soon!" + hash | print "Uh oh, trying to scrape URL again too soon!" + hash |
last_attachment_fname = doc["_attachments"].keys()[-1] | last_attachment_fname = doc["_attachments"].keys()[-1] |
last_attachment = docsdb.get_attachment(doc, last_attachment_fname) | last_attachment = docsdb.get_attachment(doc, last_attachment_fname) |
content = last_attachment | content = last_attachment |
return (doc['url'], doc['mime_type'], content.read()) | return (doc['url'], doc['mime_type'], content.read()) |
req.add_header("User-Agent", "Mozilla/4.0 (compatible; Prometheus webspider; owner maxious@lambdacomplex.org)") | 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 there is a previous version stored in couchdb, load caching helper tags |
if doc.has_key('etag'): | if doc.has_key('etag'): |
req.add_header("If-None-Match", doc['etag']) | req.add_header("If-None-Match", doc['etag']) |
if doc.has_key('last_modified'): | if doc.has_key('last_modified'): |
req.add_header("If-Modified-Since", doc['last_modified']) | req.add_header("If-Modified-Since", doc['last_modified']) |
opener = urllib2.build_opener(NotModifiedHandler()) | opener = urllib2.build_opener(NotModifiedHandler()) |
try: | try: |
url_handle = opener.open(req, None, 20) | url_handle = opener.open(req, None, 20) |
doc['url'] = url_handle.geturl() # may have followed a redirect to a new url | doc['url'] = url_handle.geturl() # may have followed a redirect to a new url |
headers = url_handle.info() # the addinfourls have the .info() too | headers = url_handle.info() # the addinfourls have the .info() too |
doc['etag'] = headers.getheader("ETag") | doc['etag'] = headers.getheader("ETag") |
doc['last_modified'] = headers.getheader("Last-Modified") | doc['last_modified'] = headers.getheader("Last-Modified") |
doc['date'] = headers.getheader("Date") | doc['date'] = headers.getheader("Date") |
doc['page_scraped'] = time.time() | doc['page_scraped'] = time.time() |
doc['web_server'] = headers.getheader("Server") | doc['web_server'] = headers.getheader("Server") |
doc['via'] = headers.getheader("Via") | doc['via'] = headers.getheader("Via") |
doc['powered_by'] = headers.getheader("X-Powered-By") | doc['powered_by'] = headers.getheader("X-Powered-By") |
doc['file_size'] = headers.getheader("Content-Length") | doc['file_size'] = headers.getheader("Content-Length") |
content_type = headers.getheader("Content-Type") | content_type = headers.getheader("Content-Type") |
if content_type != None: | if content_type != None: |
doc['mime_type'] = content_type.split(";")[0] | doc['mime_type'] = content_type.split(";")[0] |
else: | else: |
(type, encoding) = mimetypes.guess_type(url) | (type, encoding) = mimetypes.guess_type(url) |
doc['mime_type'] = type | doc['mime_type'] = type |
if hasattr(url_handle, 'code'): | if hasattr(url_handle, 'code'): |
if url_handle.code == 304: | if url_handle.code == 304: |
print "the web page has not been modified" + hash | print "the web page has not been modified" + hash |
last_attachment_fname = doc["_attachments"].keys()[-1] | last_attachment_fname = doc["_attachments"].keys()[-1] |
last_attachment = docsdb.get_attachment(doc, last_attachment_fname) | last_attachment = docsdb.get_attachment(doc, last_attachment_fname) |
content = last_attachment | content = last_attachment |
return (doc['url'], doc['mime_type'], content.read()) | return (doc['url'], doc['mime_type'], content.read()) |
else: | else: |
print "new webpage loaded" | print "new webpage loaded" |
content = url_handle.read() | content = url_handle.read() |
docsdb.save(doc) | docsdb.save(doc) |
doc = docsdb.get(hash) # need to get a _rev | doc = docsdb.get(hash) # need to get a _rev |
docsdb.put_attachment(doc, content, str(time.time()) + "-" + os.path.basename(url), doc['mime_type']) | docsdb.put_attachment(doc, content, str(time.time()) + "-" + os.path.basename(url), doc['mime_type']) |
return (doc['url'], doc['mime_type'], content) | return (doc['url'], doc['mime_type'], content) |
#store as attachment epoch-filename | #store as attachment epoch-filename |
except (urllib2.URLError, socket.timeout) as e: | except (urllib2.URLError, socket.timeout) as e: |
print "error!" | print >> sys.stderr,"error!" |
error = "" | error = "" |
if hasattr(e, 'reason'): | if hasattr(e, 'reason'): |
error = "error %s in downloading %s" % (str(e.reason), url) | error = "error %s in downloading %s" % (str(e.reason), url) |
elif hasattr(e, 'code'): | elif hasattr(e, 'code'): |
error = "error %s in downloading %s" % (e.code, url) | error = "error %s in downloading %s" % (e.code, url) |
print error | print >> sys.stderr, error |
doc['error'] = error | doc['error'] = error |
docsdb.save(doc) | docsdb.save(doc) |
return (None, None, None) | return (None, None, None) |
def scrapeAndStore(docsdb, url, depth, fieldName, agencyID): | def scrapeAndStore(docsdb, url, depth, fieldName, agencyID): |
(url, mime_type, content) = fetchURL(docsdb, url, 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"] | 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 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": | if mime_type == "text/html" or mime_type == "application/xhtml+xml" or mime_type == "application/xml": |
# http://www.crummy.com/software/BeautifulSoup/documentation.html | # http://www.crummy.com/software/BeautifulSoup/documentation.html |
soup = BeautifulSoup(content) | soup = BeautifulSoup(content) |
navIDs = soup.findAll( | navIDs = soup.findAll( |
id=re.compile('nav|Nav|menu|bar|left|right|sidebar|more-links|breadcrumb|footer|header')) | id=re.compile('nav|Nav|menu|bar|left|right|sidebar|more-links|breadcrumb|footer|header')) |
for nav in navIDs: | for nav in navIDs: |
print "Removing element", nav['id'] | print "Removing element", nav['id'] |
nav.extract() | nav.extract() |
navClasses = soup.findAll( | navClasses = soup.findAll( |
attrs={'class': re.compile('nav|menu|bar|left|right|sidebar|more-links|breadcrumb|footer|header')}) | attrs={'class': re.compile('nav|menu|bar|left|right|sidebar|more-links|breadcrumb|footer|header')}) |
for nav in navClasses: | for nav in navClasses: |
print "Removing element", nav['class'] | print "Removing element", nav['class'] |
nav.extract() | nav.extract() |
links = soup.findAll('a') # soup.findAll('a', id=re.compile("^p-")) | links = soup.findAll('a') # soup.findAll('a', id=re.compile("^p-")) |
linkurls = set([]) | linkurls = set([]) |
for link in links: | for link in links: |
if link.has_key("href"): | if link.has_key("href"): |
if link['href'].startswith("http"): | if link['href'].startswith("http"): |
# lets not do external links for now | # lets not do external links for now |
# linkurls.add(link['href']) | # linkurls.add(link['href']) |
None | None |
if link['href'].startswith("mailto"): | if link['href'].startswith("mailto"): |
# not http | # not http |
None | None |
if link['href'].startswith("javascript"): | if link['href'].startswith("javascript"): |
# not http | # not http |
None | None |
else: | else: |
# remove anchors and spaces in urls | # remove anchors and spaces in urls |
linkurls.add(fullurl(url, link['href'])) | linkurls.add(fullurl(url, link['href'])) |
for linkurl in linkurls: | for linkurl in linkurls: |
#print linkurl | #print linkurl |
scrapeAndStore(docsdb, linkurl, depth - 1, fieldName, agencyID) | scrapeAndStore(docsdb, linkurl, depth - 1, fieldName, agencyID) |
# select database | # select database |
agencydb = couch['disclosr-agencies'] | agencydb = couch['disclosr-agencies'] |
docsdb = couch['disclosr-documents'] | docsdb = couch['disclosr-documents'] |
if __name__ == "__main__": | if __name__ == "__main__": |
for row in agencydb.view('app/all'): #not recently scraped agencies view? | for row in agencydb.view('app/all'): #not recently scraped agencies view? |
agency = agencydb.get(row.id) | agency = agencydb.get(row.id) |
print agency['name'] | print agency['name'] |
for key in agency.keys(): | for key in agency.keys(): |
if key == "FOIDocumentsURL" and "status" not in agency.keys() and False: | if key == "FOIDocumentsURL" and "status" not in agency.keys() and False: |
scrapeAndStore(docsdb, agency[key], 0, key, agency['_id']) | scrapeAndStore(docsdb, agency[key], 0, key, agency['_id']) |
if key == 'website' and True: | if key == 'website' and True: |
scrapeAndStore(docsdb, agency[key], 0, key, agency['_id']) | scrapeAndStore(docsdb, agency[key], 0, key, agency['_id']) |
if "metadata" not in agency.keys(): | if "metadata" not in agency.keys(): |
agency['metadata'] = {} | agency['metadata'] = {} |
agency['metadata']['lastScraped'] = time.time() | agency['metadata']['lastScraped'] = time.time() |
if key.endswith('URL') and False: | if key.endswith('URL') and False: |
print key | print key |
depth = 1 | depth = 1 |
if 'scrapeDepth' in agency.keys(): | if 'scrapeDepth' in agency.keys(): |
depth = agency['scrapeDepth'] | depth = agency['scrapeDepth'] |
scrapeAndStore(docsdb, agency[key], depth, key, agency['_id']) | scrapeAndStore(docsdb, agency[key], depth, key, agency['_id']) |
agencydb.save(agency) | agencydb.save(agency) |
import sys,os | import sys,os |
sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../')) | sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../')) |
import genericScrapers | import genericScrapers |
import scrape | import scrape |
from bs4 import BeautifulSoup | from bs4 import BeautifulSoup |
#http://www.doughellmann.com/PyMOTW/abc/ | #http://www.doughellmann.com/PyMOTW/abc/ |
class ScraperImplementation(genericScrapers.GenericOAICDisclogScraper): | class ScraperImplementation(genericScrapers.GenericOAICDisclogScraper): |
def getDescription(self,content, entry,doc): | def getDescription(self,content, entry,doc): |
link = None | link = None |
links = [] | links = [] |
description = "" | description = "" |
for atag in entry.find_all('a'): | for atag in entry.find_all('a'): |
if atag.has_key('href'): | if atag.has_key('href'): |
link = scrape.fullurl(self.getURL(),atag['href']) | link = scrape.fullurl(self.getURL(),atag['href']) |
(url,mime_type,htcontent) = scrape.fetchURL(scrape.docsdb, link, "foidocuments", self.getAgencyID(), False) | (url,mime_type,htcontent) = scrape.fetchURL(scrape.docsdb, link, "foidocuments", self.getAgencyID(), False) |
if htcontent != None: | if htcontent != None: |
if mime_type == "text/html" or mime_type == "application/xhtml+xml" or mime_type =="application/xml": | if mime_type == "text/html" or mime_type == "application/xhtml+xml" or mime_type =="application/xml": |
# http://www.crummy.com/software/BeautifulSoup/documentation.html | # http://www.crummy.com/software/BeautifulSoup/documentation.html |
soup = BeautifulSoup(htcontent) | soup = BeautifulSoup(htcontent) |
for row in soup.find(class_ = "ms-rteTable-GreyAlternating").find_all('tr'): | rowtitle = soup.find(class_ = "wc-title").find("h1").string |
if rowtitle != None: | |
description = rowtitle + ": " | |
for row in soup.find(class_ ="wc-content").find_all('td'): | |
if row != None: | if row != None: |
rowtitle = row.find('th').string | for text in row.stripped_strings: |
if rowtitle != None: | description = description + text + "\n" |
description = description + "\n" + rowtitle + ": " | |
for text in row.find('td').stripped_strings: | |
description = description + text | |
for atag in row.find_all("a"): | for atag in row.find_all("a"): |
if atag.has_key('href'): | if atag.has_key('href'): |
links.append(scrape.fullurl(link,atag['href'])) | links.append(scrape.fullurl(link,atag['href'])) |
if links != []: | if links != []: |
doc.update({'links': links}) | doc.update({'links': links}) |
if description != "": | if description != "": |
doc.update({ 'description': description}) | doc.update({ 'description': description}) |
def getColumnCount(self): | def getColumnCount(self): |
return 2 | return 2 |
def getTable(self,soup): | def getTable(self,soup): |
return soup.find(class_ = "ms-rteTable-GreyAlternating") | return soup.find(class_ = "ms-rteTable-default") |
def getColumns(self,columns): | def getColumns(self,columns): |
(date, title) = columns | (date, title) = columns |
return (title, date, title, title, None) | return (title, date, title, title, None) |
if __name__ == '__main__': | if __name__ == '__main__': |
print 'Subclass:', issubclass(ScraperImplementation, genericScrapers.GenericOAICDisclogScraper) | print 'Subclass:', issubclass(ScraperImplementation, genericScrapers.GenericOAICDisclogScraper) |
print 'Instance:', isinstance(ScraperImplementation(), genericScrapers.GenericOAICDisclogScraper) | print 'Instance:', isinstance(ScraperImplementation(), genericScrapers.GenericOAICDisclogScraper) |
ScraperImplementation().doScrape() | ScraperImplementation().doScrape() |
import sys,os | import sys,os |
sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../')) | sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../')) |
import genericScrapers | import genericScrapers |
import scrape | import scrape |
from bs4 import BeautifulSoup | from bs4 import BeautifulSoup |
#http://www.doughellmann.com/PyMOTW/abc/ | #http://www.doughellmann.com/PyMOTW/abc/ |
class ScraperImplementation(genericScrapers.GenericOAICDisclogScraper): | class ScraperImplementation(genericScrapers.GenericOAICDisclogScraper): |
def getColumns(self,columns): | def getColumns(self,columns): |
(id, date, title, description, notes) = columns | (id, date, title, description, notes) = columns |
return (id, date, title, description, notes) | return (id, date, title, description, notes) |
def getTable(self,soup): | def getTable(self,soup): |
return soup.find(id = "content").table | return soup.find("table") |
if __name__ == '__main__': | if __name__ == '__main__': |
print 'Subclass:', issubclass(ScraperImplementation, genericScrapers.GenericOAICDisclogScraper) | print 'Subclass:', issubclass(ScraperImplementation, genericScrapers.GenericOAICDisclogScraper) |
print 'Instance:', isinstance(ScraperImplementation(), genericScrapers.GenericOAICDisclogScraper) | print 'Instance:', isinstance(ScraperImplementation(), genericScrapers.GenericOAICDisclogScraper) |
ScraperImplementation().doScrape() | ScraperImplementation().doScrape() |