[kepler-dev] Rearranging Kepler jars

Dan Higgins higgins at nceas.ucsb.edu
Mon May 14 15:50:26 PDT 2007

Hi All,
    As anyone who has looked into the$Kepler/lib.jar directory knows, 
Kepler has a very large number of jar files. In an attempt to sort out 
just what jars get used where, I experimented to find out just what jars 
were needed for starting up Kepler if one did not load the actor code at 
startup. I thus removed all the actor kar files (but for a single 
Ptolemy actor that called no additonal java code). I also tried doing an 
ecogrid data search to find jars used there. The result is a set of 29 
jars that are needed for starting up Kepler and searching for data. I am 
calling this set of jars 'base-jars' and they are now in the 
'$Kepler/lib/jar/base-jars/' directory in the head of CVS. All jars NOT 
in this directory are assumed to be needed for specific actors.

    Eventually, we hope to load actor specific jars with actor kar 
files. This rearrangement of jars is just a first step in trying to 
simplify the problem of a having a confusingly large number of jar files.

    Note that the 'actor specific jars' have not been moved, and 
(hopefully) building Kepler should work the same as before. I did try 
putting the actor-specific jars in their own directory, but a problem 
occurs with the cipres jars on launch - apparently the cipres code 
requires the cipres jars to be in their current location!


Dan Higgins                                  higgins at nceas.ucsb.edu
http://www.nceas.ucsb.edu/    Ph: 805-893-5127
National Center for Ecological Analysis and Synthesis (NCEAS) Marine Science Building - Room 3405
Santa Barbara, CA 93195

More information about the Kepler-dev mailing list