In this case package would be enough. If you run a cluster the assembly may be required.
On Fri, Oct 25, 2013 at 3:08 PM, Umar Javed <[email protected]> wrote: > I'm sorry I didn't quite understand that. I'm using and running spark on > my local machine. What did you mean by infrastructure? > > > On Fri, Oct 25, 2013 at 4:01 AM, Sergey Soldatov < > [email protected]> wrote: > >> For building Spark only you may use sbt/sbt package. It's much faster. >> But if you use Spark in some infrastructure which requires the assembly, >> the only way is to use assembly target. >> >> >> On Fri, Oct 25, 2013 at 2:34 PM, Umar Javed <[email protected]>wrote: >> >>> thanks. This takes a lot of time though. It takes me 10 mins for the >>> build to finish after changing a single line of code. Is there a quicker >>> way? >>> >>> >>> On Fri, Oct 25, 2013 at 2:35 AM, Sergey Soldatov < >>> [email protected]> wrote: >>> >>>> Hi Umar, >>>> Exactly. You need to use sbt/sbt assembly >>>> It will compile only changed files and build the assembly. >>>> >>>> Thanks, >>>> Sergey >>>> >>>> >>>> On Fri, Oct 25, 2013 at 1:29 PM, Umar Javed <[email protected]>wrote: >>>> >>>>> Hi, >>>>> >>>>> After making a change to a .scala file in the spark source code, how I >>>>> build it using sbt? Do I have to use 'sbt/sbt assembly' again? >>>>> >>>>> cheers, >>>>> Umar >>>>> >>>> >>>> >>> >> >
