I don't know yet. The actual implementation (they call it python_legacy) works. But I think it only supports Amazon S3. I'm using Azure, so I needed to add that support. That's the only thing I added. So someone using S3 could use the library as-is. They are re-writing the python library to something more pythonic and also more inline with the feature set of the Java API. With the new implementation, they told me that it would be easier to support other connectors like Azure and HadoopCatalog would also be better supported.