This project provides a serializer and raw comparator for using TBase and TEnum objects in Hadoop jobs. It works seamlessly with Cascading and Cascalog.
To make use of this serializer, add "backtype.hadoop.ThriftSerialization" to the list of serializers listed under "io.serializations" in your JobConf. Then you can use Thrift objects directly in your tuples.
Cascading-Thrift supports both Cascading 1.2 and Cascading 2.0.
For Cascading 1.2 support, use
[backtype/cascading-thrift "0.1.0"]
For Cascading 2.0, use
[backtype/cascading-thrift "0.2.5"]