Build System Use Cases

STATUS:

The content on this page is outdated. The page is archived for reference only. For more information about current work, please contact the Build System Team.

Overview

Safely revising the build system as part of the Kepler/CORE effort requires understanding the functions of the current system. The purpose of this document is not to capture every detail about how the current system works, but rather to model the key functions (at the level of what are sometimes called 'business' use cases) so that (a) the functions needed in the future can be maintained, and (b) the current functions can be compared with those functions proposed for the revised build system.

The following captures what the current build system does from the point of view of a Kepler developer. Note that we treat the variously optimized implementations of particular tasks as the same business use case.

Comments, corrections, additions and clarifications are welcome!

Use cases

COMPILING

• Compile portion of Ptolemy used by Kepler.
• Compile a local copy of the Kepler system.

PACKAGING

• Build all kar files describing component to be included in the library at run time.
• Build one kar file.
• Package compiled code and resources in jars.

RUNNING

• Run Ptolemy using the Vergil user interface.
• Start the Kepler GUI with a blank canvas.
• Start the Kepler GUI, pre-loading a specific workflow.
• Start Kepler with JINI support enabled.
• Run a workflow from the command line with no support for graphical (e.g., Swing) components.
• Run a workflow from the command line with support for graphical components enabled.

TESTING

• Run a set of unit tests.
• Run one unit test.
• Run a set of workflow tests.

DOCUMENTATION

• Build documentation.

DISTRIBUTION

• Create Windows batch and Unix script files to start the Kepler GUI.
• Create Windows batch and Unix script files to run multiple workflows from the command line.
• Create Windows batch and Unix script files to run one workflow from the command line (both with and without graphics support).
• Create a Java Web Start distribution of Kepler.
• Sign all jars for use with Java Web Start.
• Create a zip archive of the Kepler source code.
• Build a Kepler WAR file.
• Build Windows executables.
• Run Mac JarBundler to create an executable jar for Mac.
• Run IzPack to create an installer for Kepler.

MISCELLANEOUS

• Check out necessary (run-time) Eco Grid jars from Eco Grid repository so that these can be included in the build.
• Start a Kepler repository instance and add kar files to repository.
• Run the Kepler QueryBuilder.
• Run the Kepler DataCacheViewer.
• Retrieve EML files from EML module in the Ecoinformatics CVS repository.
• Run Find Bugs Program on Kepler Source.

Contributors to this document

• David Welker
• Tim McPhillips