Merge pull request #475 from coreos-inc/seofix

Use a proper HTML parser with BS and catch exceptions

util/seo.py

@@ -32,11 +32,13 @@ def render_snapshot(url):
   # Remove script tags
   logger.info('Removing script tags: %s' % url)
 
-  soup = BeautifulSoup(out_html.decode('utf8'))
-  to_extract = soup.findAll('script')
-  for item in to_extract:
-    item.extract()
-
-  logger.info('Snapshotted url: %s' % url)
+  try:
+    soup = BeautifulSoup(out_html.decode('utf8'), 'html.parser')
+    to_extract = soup.findAll('script')
+    for item in to_extract:
+      item.extract()
+  except:
+    logger.exception('Exception when trying to parse served HTML')
+    return out_html
 
   return str(soup)