This document describes what tools and third-party software are required to build the CLRA application, as well as how to download the source code for the application.
Back to Top
The CRLA project requires SSH and CVS. These applications come pre-installed with many Linux flavors (e.g. Redhat). Windows users will have to install these applications; see Using CVS (with Windows).Back to Top
The CLRA site is written in Java using open-source software (with one small caveat, which is that Oracle is an optional substitute for MySQL as a database. Oracle is good for development, because it allows one to add foreign-key and other constraints on data relationships. MySQL doesn't have constraints or triggers, but MySQL doesn't require a commercial license for production use.) Here is what you'll need for development:
Recommended: Linux with the 2.4.2 kernel
Other: Windows NT 4.0 (SP 6), Windows 2000, or Windows XP
Not recommended: Windows ME, 98 or 95.
The JDK is available from Sun Microsystems: http://java.sun.com/j2se/.
The J2EE developer's kit is available from Sun Microsystems: http://java.sun.com/j2ee/download.html.
See the Tools section above. SSH is required to check code into and out of SourceForge. SSH is basically a secure form of Telnet.
SourceForge has a nice intro to SSH:
Accounts are free, but registration is required:
MySQL is easy to install on both Windows and Linux. It is available from www.mysql.com
The developer copy of Oracle is full featured, but can't be used for production. Oracle is available for download from Oracle's site. Oracle is easy to install on Windows, but version 8.17 is difficult to install on Linux.
JBoss is available from www.jboss.org.
Struts is available from jakarta.apache.org/struts.
Log4j is available from www.junit.org.
The CLRA source code is available either as release bundles or as CVS snapshots.
Release bundles are listed under the "Files" section of the CLRA project site.
Directions for obtaining CVS snapshots are available under the "CVS" section. The module that you'll need is called "clra-java".
One trick with CVS: after an initial checkout, use the "cvs update -dP clra-java" command to prune dead directories from the source code tree.
SourceForge has some nice documentation on using CVS with SourceForge. See the "6. CVS - general" and "7. CVS - OS specific" topics on Site Docs section of SourceForge.
However you obtain the CLRA source code, once it is installed on your machine, you should find a directory structure which matches the one described in CLRA files and directories.Back to Top
You may find it convenient to set up an environment variable that points to the top of the clra-java source code tree. This environment variable is not required by the build scripts, but it is a convenient way to refer to the source code directory in the documentation that follows.
When properly defined, $CLRA_HOME/bin points to the script-file directory; $CLRA_HOME/src points to the source-code directory; and so on.Back to Top
A source-code drop contains everything necessary to build, but not deploy, the CLRA application. To check whether your environment is configured properly to this point, you may wish to try a test build. The following command, issued from $CLRA_HOME, will clean up any previous build and compile the source code without deploying the CLRA application:
./build.sh clean build.core
For more details on building the CLRA application, see the documentation on the CLRA build process.Back to Top
Now that third-party software and the CLRA source-code is installed on your machine, the next step is to set up a database for the CLRA application. Database configuration is covered in a separate document, Setting up the CLRA database.Back to Top