So, as far as I can see, Hadoop has the main developer area for core Hadoop code, unit tests in the test directories, user scripts (like hadoop/mapred/yarn), and build scripts.
I've got some utilities that are really for Hadoop contributors. These serve two purposes: 1. These are just generally useful as private API examples 2. They have some utility for developer purposes (e.g. the random .jhist generator I'm working on for MAPREDUCE-6376) Does anyone have suggestions for where such code bits (and possibly corresponding scripts) should go? -Ray
