Throw some checking aroud the build_doc

mitechie · mitechie · commit 2ee2fe953675 · 2012-04-18T15:02:01.000-04:00
diff --git a/src/readability_lxml/encoding.py b/src/readability_lxml/encoding.py
@@ -7,8 +7,6 @@
 
 
 def get_encoding(page):
-    LOG.info('GET ENCODING')
-    LOG.info(type(page))
     text = re.sub('</?[^>]*>\s*', ' ', page)
     enc = 'utf-8'
     if not text.strip() or len(text) < 10:
diff --git a/src/readability_lxml/htmls.py b/src/readability_lxml/htmls.py
@@ -17,8 +17,10 @@
 
 
 def build_doc(page):
-    LOG.info('BUILD DOC')
-    LOG.info(type(page))
+    """Requires that the `page` not be None"""
+    if page is None:
+        LOG.error("Page content is None, can't build_doc")
+        return ''
     if isinstance(page, unicode):
         page_unicode = page
     else:
