- archives have hashes too

oilarchive/model.py

@@ -51,7 +51,7 @@ class ArchiveItem(SQLObject):
   authorSort = UnicodeCol(default = '')
   description = UnicodeCol() # text(for rss)
   html = UnicodeCol() #(for page, contains javascript)
-  text = UnicodeCol() #Fulltext 
+  text = UnicodeCol(length = 2**25) #Fulltext 
   relDate = DateTimeCol() #timestamp (item released)
   pubDate = DateTimeCol() #timestamp (item published)
   modDate = DateTimeCol() #timestamp (item published)
@@ -137,6 +137,9 @@ class Archive(SQLObject):
   
   hashId = UnicodeCol(alternateID = True, length=128)
   
+  def setHashId(self):
+    self.hashId = md5.new("%s" % self.id).hexdigest()
+    
   def _get_pubDateTimestamp(self):
     if self.initialized:
       return int(time.mktime(self.pubDate.timetuple()))
@@ -168,19 +171,24 @@ class Archive(SQLObject):
     
   def update(self):
     result = simplejson.loads(read_url(self.files_url))
-    if result.has_key('css'):
+    if result and result.has_key('css'):
       self.css = read_url(self.full_url(result['css']))
     else:
       self.css = ''
-    if result.has_key('js'):
+    if result and result.has_key('js'):
       self.js = read_url(self.full_url(result['js']))
     else:
       self.js = ''
     result = simplejson.loads(read_url(self.update_url))
     items = result.get('items', [])
+    print len(items)
     for id in items:
-      print "updating / adding ", id
-      data = jsonLoadArchiveItem(read_url(self.data_url(id)))
+      try:
+        data = read_url(self.data_url(id))
+        data = jsonLoadArchiveItem(data)
+      except:
+        print "failed to load ", id, "from ", self.data_url(id)
+        continue
       q = ArchiveItem.select(AND(
             ArchiveItem.q.archiveItemId == id, 
             ArchiveItem.q.archiveID == self.id))