Refactoring the Mahout Shell

[Trevor Grant <tr...@gmail.com>]

Hey all,

The Mahout shell has caused issues with upgrades in the past.  As I was
trying to migrate Mahout to Spark 2.x- the shell issue crept up again.
There were many chances to the way the shell operates in Spark.

So my work around was to simply drop the MahoutILoop class / Mahout module:
spark-shell, in favor of using the spark-shell that comes with the spark
distribution, and pass a script to it which loads Mahout libraries and sets
up a Mahout Distributed Context.

To be clear, $MAHOUT_HOME/bin/mahout spark-shell will still launch a Mahout
enabled spark shell as it has, the difference is how it will do that.

It might make more sense to simply look at the code:
https://github.com/apache/mahout/pull/271

and especially:
https://github.com/apache/mahout/pull/271/files#diff-21970d9ce94858765ec000a3d091e649R294


I want to float this idea on dev and see if there are any objections or
gotchyas to going this route.

Thanks!


Trevor Grant
Data Scientist
https://github.com/rawkintrevo
http://stackexchange.com/users/3002022/rawkintrevo
http://trevorgrant.org

*"Fortunate is he, who is able to know the causes of things."  -Virgil*