From: maxious <maxious@lambdacomplex.org>
Date: Mon, 30 Jan 2012 09:18:35 +0000
Subject: Store scraper results to couchdb
X-Git-Url: https://maxious.lambdacomplex.org/git/?p=disclosr.git&a=commitdiff&h=d65e3708eccd07bb0a5284b8d2c6effd10dd75c8
---
Store scraper results to couchdb


Former-commit-id: 234bb19e5682c98cb4cbd9c6d6b1bf542ff16d50
---


--- a/scrape.py
+++ b/scrape.py
@@ -4,6 +4,9 @@
 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):  
@@ -12,11 +15,13 @@
         addinfourl.code = code
         return addinfourl
 
-def scrapeAndStore(docsdb, url, depth, agencyID):
+def fetchURL(docsdb, url, agencyID):
     hash = hashlib.md5(url).hexdigest()
     req = urllib2.Request(url)
     print "Fetching %s", url
-    doc = docsdb['hash']
+    doc = docsdb.get(hash) 
+    if doc == None:
+	doc = {'_id': hash, 'agencyID': agencyID}
     #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'])
@@ -29,44 +34,59 @@
     doc['etag'] = headers.getheader("ETag")
     doc['last_modified'] = headers.getheader("Last-Modified") 
     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") 
-     
+    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
         else: 
-            #do scraping
-            html = url_handle.read()
+            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
+        #record/alert error to error database
+    
+
+
+
+def scrapeAndStore(docsdb, url, depth, agencyID):
+    (mime_type,content) = fetchURL(docsdb, url, agencyID)
+    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(html)
-        links = soup.findAll('a') # soup.findAll('a', id=re.compile("^p-"))
-        for link in links:
-            if link.has_key("href"):
-                print link['href']
+            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-"))
+            for link in links:
+            	if link.has_key("href"):
+			if link['href'].startswith("http"):
+				linkurl = link['href']
+			else:
+                		linkurl = urljoin(url,link['href'])
+			print linkurl
                 #for each unique link
+		#	if 
                 #if html mimetype
                 # go down X levels,
                 # diff with last stored attachment, store in document
                 #if not
                 #   remember to save parentURL and title (link text that lead to document)
     
-    #store as attachment epoch-filename
-    else:
-        print "error %s in downloading %s", url_handle.code, URL
-        #record/alert error to error database
-    
-    
 
-
-
-
-
-
-
-
-
-couch = couchdb.Server('http://192.168.1.148:5984/')
+couch = couchdb.Server('http://127.0.0.1:5984/')
 
 # select database
 agencydb = couch['disclosr-agencies']