Extracting Reuters 21578 dataset

 Hello 
I have a small problem , when i try to extract Reuters21578 dataset using OpenCalais in wandora. I'm getting following error
"This document has errors that must be fixed before using HTML Tidy to generate a tidied up version"
Here is the link for the reuters dataset
http://www.daviddlewis.com/resources/testcollections/reuters21578/
Can OpenCalais team suggest how can i fix this issue?
 
Thanks