From: Maxious Date: Mon, 03 Dec 2012 09:22:28 +0000 Subject: pdf scrapers X-Git-Url: http://maxious.lambdacomplex.org/git/?p=disclosr.git&a=commitdiff&h=8c65f1f2248f9f6748c09a69638e43e9389888be --- 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 @@ -15,10 +15,6 @@ from StringIO import StringIO -from docx import * -from lxml import etree -import zipfile - from pdfminer.pdfparser import PDFDocument, PDFParser from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter, process_pdf from pdfminer.pdfdevice import PDFDevice, TagExtractor @@ -39,14 +35,14 @@ """ disclosr agency id """ if self.agencyID is None: self.agencyID = os.path.basename(sys.argv[0]).replace(".py", "") - return self.agencyID + 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 + return self.disclogURL @abc.abstractmethod def doScrape(self): @@ -62,14 +58,15 @@ self.getURL(), "foidocuments", self.getAgencyID()) laparams = LAParams() rsrcmgr = PDFResourceManager(caching=True) - outfp = StringIO.StringIO() + outfp = StringIO() device = TextConverter(rsrcmgr, outfp, codec='utf-8', laparams=laparams) - fp = StringIO.StringIO() - fp.write(content) - description = output.getvalue() + 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() @@ -77,11 +74,10 @@ doc = foidocsdb.get(dochash) if doc is None: print "saving " + dochash - edate = datetime.fromtimestamp(mktime()).strftime("%Y-%m-%d") + edate = date.today().strftime("%Y-%m-%d") doc = {'_id': dochash, 'agencyID': self.getAgencyID() , 'url': self.getURL(), 'docID': dochash, - "date": edate, "title": "Disclosure Log Updated"} - self.getDescription(entry, entry, doc) + "date": edate, "title": "Disclosure Log Updated", "description": description} foidocsdb.save(doc) else: print "already saved" @@ -103,17 +99,16 @@ 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) + 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 = datetime.fromtimestamp(mktime()).strftime("%Y-%m-%d") + edate = time().strftime("%Y-%m-%d") doc = {'_id': dochash, 'agencyID': self.getAgencyID() , 'url': self.getURL(), 'docID': dochash, - "date": edate, "title": "Disclosure Log Updated"} - self.getDescription(entry, entry, doc) + "date": edate, "title": "Disclosure Log Updated", "description": description} foidocsdb.save(doc) else: print "already saved" --- 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 external links for now - # linkurls.add(link['href']) - None - if link['href'].startswith("mailto"): - # not http - None - if link['href'].startswith("javascript"): - # not http - None - else: - # remove anchors and spaces in urls - linkurls.add(fullurl(url,link['href'])) - for linkurl in linkurls: - #print linkurl - scrapeAndStore(docsdb, linkurl, depth-1, 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 external links for now + # linkurls.add(link['href']) + None + if link['href'].startswith("mailto"): + # not http + None + if link['href'].startswith("javascript"): + # not http + None + else: + # remove anchors and spaces in urls + linkurls.add(fullurl(url,link['href'])) + for linkurl in linkurls: + #print linkurl + scrapeAndStore(docsdb, linkurl, depth-1, fieldName, agencyID) #couch = couchdb.Server('http://192.168.1.148:5984/') couch = couchdb.Server('http://127.0.0.1:5984/') @@ -196,20 +198,20 @@ docsdb = couch['disclosr-documents'] if __name__ == "__main__": - for row in agencydb.view('app/getScrapeRequired'): #not recently scraped agencies view? - agency = agencydb.get(row.id) - print agency['name'] - for key in agency.keys(): - if key == "FOIDocumentsURL" and "status" not in agency.keys: - scrapeAndStore(docsdb, agency[key],0,key,agency['_id']) - if key == 'website' and False: - scrapeAndStore(docsdb, agency[key],0,key,agency['_id']) - agency['metadata']['lastScraped'] = time.time() - if key.endswith('URL') and False: - print key - depth = 1 - if 'scrapeDepth' in agency.keys(): - depth = agency['scrapeDepth'] - scrapeAndStore(docsdb, agency[key],depth,key,agency['_id']) - agencydb.save(agency) - + for row in agencydb.view('app/getScrapeRequired'): #not recently scraped agencies view? + agency = agencydb.get(row.id) + print agency['name'] + for key in agency.keys(): + if key == "FOIDocumentsURL" and "status" not in agency.keys: + scrapeAndStore(docsdb, agency[key],0,key,agency['_id']) + if key == 'website' and False: + scrapeAndStore(docsdb, agency[key],0,key,agency['_id']) + agency['metadata']['lastScraped'] = time.time() + if key.endswith('URL') and False: + print key + depth = 1 + if 'scrapeDepth' in agency.keys(): + depth = agency['scrapeDepth'] + scrapeAndStore(docsdb, agency[key],depth,key,agency['_id']) + agencydb.save(agency) + --- /dev/null +++ b/documents/scrapers/0049d35216493c545ef5f7f000e6b252.py @@ -1,1 +1,19 @@ +import sys +import os +sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../')) +import genericScrapers + +class ScraperImplementation(genericScrapers.GenericPDFDisclogScraper): + + def __init__(self): + super(ScraperImplementation, self).__init__() + + +if __name__ == '__main__': + print 'Subclass:', issubclass(ScraperImplementation, + genericScrapers.GenericOAICDisclogScraper) + print 'Instance:', isinstance(ScraperImplementation(), + genericScrapers.GenericOAICDisclogScraper) + ScraperImplementation().doScrape() + --- a/documents/scrapers/0049d35216493c545ef5f7f000e6b252.txt +++ /dev/null @@ -1,2 +1,1 @@ -pdf --- /dev/null +++ b/documents/scrapers/6afdde1d4ff1ad8d8cfe1a8675ea83bd.py @@ -1,1 +1,19 @@ +import sys +import os +sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../')) +import genericScrapers + +class ScraperImplementation(genericScrapers.GenericPDFDisclogScraper): + + def __init__(self): + super(ScraperImplementation, self).__init__() + + +if __name__ == '__main__': + print 'Subclass:', issubclass(ScraperImplementation, + genericScrapers.GenericOAICDisclogScraper) + print 'Instance:', isinstance(ScraperImplementation(), + genericScrapers.GenericOAICDisclogScraper) + ScraperImplementation().doScrape() + --- a/documents/scrapers/6afdde1d4ff1ad8d8cfe1a8675ea83bd.txt +++ /dev/null @@ -1,2 +1,1 @@ -PDF --- /dev/null +++ b/documents/scrapers/8317df630946937864d31a4728ad8ee8.py @@ -1,1 +1,19 @@ +import sys +import os +sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../')) +import genericScrapers + +class ScraperImplementation(genericScrapers.GenericPDFDisclogScraper): + + def __init__(self): + super(ScraperImplementation, self).__init__() + + +if __name__ == '__main__': + print 'Subclass:', issubclass(ScraperImplementation, + genericScrapers.GenericOAICDisclogScraper) + print 'Instance:', isinstance(ScraperImplementation(), + genericScrapers.GenericOAICDisclogScraper) + ScraperImplementation().doScrape() + --- a/documents/scrapers/8317df630946937864d31a4728ad8ee8.txt +++ /dev/null @@ -1,2 +1,1 @@ -pdf --- /dev/null +++ b/documents/scrapers/c57c0bf315ce5977e730905707a2f6a3.py @@ -1,1 +1,19 @@ +import sys +import os +sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../')) +import genericScrapers + +class ScraperImplementation(genericScrapers.GenericPDFDisclogScraper): + + def __init__(self): + super(ScraperImplementation, self).__init__() + + +if __name__ == '__main__': + print 'Subclass:', issubclass(ScraperImplementation, + genericScrapers.GenericOAICDisclogScraper) + print 'Instance:', isinstance(ScraperImplementation(), + genericScrapers.GenericOAICDisclogScraper) + ScraperImplementation().doScrape() + --- a/documents/scrapers/c57c0bf315ce5977e730905707a2f6a3.txt +++ /dev/null @@ -1,3 +1,1 @@ -# pdf -http://www.awm.gov.au/about/AWM_Disclosure_Log.pdf --- a/documents/template.inc.php +++ b/documents/template.inc.php @@ -150,10 +150,10 @@ $result = ""; $result .= '
'; $result .= '

: " . truncate($row->value->title, 120) . ""; - $result .= '(' . $idtoname[$row->value->agencyID] . ')

'; - $result .= "

Title" . $row->value->title . "
"; + $result .= ' (' . $idtoname[$row->value->agencyID] . ')'; + $result .= "

Title: " . $row->value->title . "
"; if (isset($row->value->description)) { - $result .= str_replace("\n", "
", $row->value->description); + $result .= str_replace("\n", "
", preg_replace("/(^[\r\n]*|[\r\n]+)[\s\t]*[\r\n]+/", "",trim($row->value->description))); } if (isset($row->value->notes)) { $result .= "
Note: " . $row->value->notes;