Home

cascading.jruby is a DSL for Cascading, which is a dataflow API written in Java. With cascading.jruby, Ruby programmers can rapidly script efficient MapReduce jobs for Hadoop.

cascading.jruby provides a clean Ruby interface to Cascading, but doesn't attempt to add abstractions on top of it. Therefore, you should be acquainted with the Cascading API before you begin.

For operations you can apply to your dataflow within a pipe assembly, see the Assembly class. For operations available within a block passed to a group_by, union, or join, see the Aggregations class.

Note that the Ruby code you write merely constructs a Cascading job, so no JRuby runtime is required on your cluster. This stands in contrast with writing Hadoop streaming jobs in Ruby. To run cascading.jruby applications on a Hadoop cluster, you must use Jading to package them into a job jar.

cascading.jruby has been tested on JRuby versions 1.2.0, 1.4.0, 1.5.3, 1.6.5, and 1.6.7.2.

If you're new to cascading.jruby, see Getting Started.

If you want to contribute to cascading.jruby, see Hacking.

For an overview of the previous releases of cascading.jruby, see the History.