|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 | #http://packages.python.org/CouchDB/client.html import couchdb import urllib2 from BeautifulSoup import BeautifulSoup import re import hashlib from urlparse import urljoin import time import os #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 def fetchURL(docsdb, url, fieldName, agencyID, scrape_again=True): hash = hashlib.md5(url).hexdigest() req = urllib2.Request(url) print "Fetching %s" % url doc = docsdb.get(hash) if doc == None: doc = {'_id': hash, 'agencyID': agencyID, 'url': url, 'fieldName':fieldName} else: if (time.time() - doc['page_scraped']) < 3600: 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['mime_type'],last_attachment) if scrape_again == False: print "Not scraping this URL again as requested" return (None,None) time.sleep(3) # wait 3 seconds to give webserver time to recover #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()) url_handle = opener.open(req) 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['powered_by'] = headers.getheader("X-Powered-By") doc['file_size'] = headers.getheader("Content-Length") doc['mime_type'] = headers.getheader("Content-Type").split(";")[0] if hasattr(url_handle, 'code'): if url_handle.code == 304: print "the web page has not been modified" return (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['mime_type'], content) #store as attachment epoch-filename else: print "error %s in downloading %s" % url_handle.code, URL doc['error'] = "error %s in downloading %s" % url_handle.code, URL docsdb.save(doc) return (None,None) def scrapeAndStore(docsdb, url, depth, fieldName, agencyID): (mime_type,content) = fetchURL(docsdb, url, fieldName, agencyID) if content != None and depth > 0: 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')) for nav in navIDs: print "Removing element", nav['id'] nav.extract() navClasses = soup.findAll(attrs={'class' : re.compile('nav|menu|bar')}) 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 else: linkurls.add(urljoin(url,link['href'].replace(" ","%20"))) for linkurl in linkurls: #print linkurl scrapeAndStore(docsdb, linkurl, depth-1, fieldName, agencyID) couch = couchdb.Server('http://127.0.0.1:5984/') # select database agencydb = couch['disclosr-agencies'] docsdb = couch['disclosr-documents'] 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 == 'website' or key.endswith('URL'): print key scrapeAndStore(docsdb, agency[key],agency['scrapeDepth'],key,agency['_id']) agency['metadata']['lastscraped'] = time.time() agencydb.save(agency) |