Merge pull request #742 from Gluejar/fix-scraper

fix can_scrape

core/loaders/scrape.py

@@ -22,7 +22,7 @@ class BaseScraper(object):
     designed to make at least a decent guess for webpages that embed metadata
     '''
     can_scrape_hosts = False
-    can_scrape_strings = [''] #should always return true
+    can_scrape_strings = False 
     @classmethod
     def can_scrape(cls, url):
         ''' return True if the class can scrape the URL '''

core/loaders/ubiquity.py

@@ -22,7 +22,7 @@ class UbiquityScraper(BaseScraper):
 
     def get_language(self):
         langlabel = self.doc.find(string='Language')
-        lang = langlabel.parent.parent.find_next_sibling()
+        lang = langlabel.parent.parent.find_next_sibling() if langlabel else ''
         lang = lang.get_text() if lang else ''
         lang = get_language_code(lang) if lang else ''
         if lang: