Merge pull request buriy#56 from nathanathan/patch-1

buriy · buriy · commit 2d4cfdb2c8e5 · 2014-12-20T02:11:53.000+05:00
Defaulting to utf-8 when chardet returns None
diff --git a/readability/encoding.py b/readability/encoding.py
@@ -26,7 +26,7 @@ def get_encoding(page):
     if not text.strip() or len(text) < 10:
         return enc # can't guess
     res = chardet.detect(text)
-    enc = res['encoding']
+    enc = res['encoding'] or 'utf-8'
     #print '->', enc, "%.2f" % res['confidence']
     enc = custom_decode(enc)
     return enc
@@ -45,4 +45,4 @@ def custom_decode(encoding):
     if encoding in alternates:
         return alternates[encoding]
     else:
-        return encoding
+        return encoding
