Yeah, the intent was to reuse all the code in AbstractJob. This splits up the support. I'd prefer to go all the way if you're doing this and keep all the job standardization in one place.
On Sat, Nov 5, 2011 at 12:42 AM, Grant Ingersoll <[email protected]> wrote: > I've moved some of the AbstractJob.prepareJob functionality over to statics > in HadoopUtil so that non AbstractJob's can take advantage. The calls in > AbstractJob now use HadoopUtil. > > We have a ton of places where we can replace a lot of repetitive code with > HadoopUtil.prepareJob(). > > Cheers, > Grant
