JarAnalyzer

JarAnalyzer was developed to be a dependency management utility for jar files. It's primary purpose is to traverse through a directory, parse each of the jar files in that directory, and identify the dependencies between the jar files. The output is an xml file representing the PhysicalDependencies between the jar files.

Running JarAnalyzer is easy. Download the binary version, unzip the file, and execute the runxmlsummary.bat or rundotsummary.bat file. Type in the name of the directory with the jar files you want to analyze. Next, type in the name of the output .xml or .grph file.

You can also specify filters to exclude certain packages in jar files when examining the dependencies. This will help minimize the number of unresolved dependencies on packages that you are not interested in reporting on. The default filter in Filter.properties is to exclude all packages beginning with java. and javax. You can also choose to filter specific jar files.

The output is an xml file of the dependencies between the jar files in the directory specified when JarAnalyzer was run. Here's a listing of the primary elements that JarAnalyzer reports on. - JarName: The name of the jar file analyzed. - Statistics: Number of packages and classes in the jar file. Inner class are included in this count. The Level represents where in the hierarchy a jar file resides. Level 1 components are at the bottom. Level 2 depend on at least one Level 1. Level 3 depend on at least one Level 2. And so on. - Metrics: These are based on the "Martin Metrics". - Packages: Names of the packages in the jar file. - OutgoingDependencies: Dependent jar files based on what was found in the directory being analyzed. - IncomingDependencies: Jar files dependent on this jar file. - Cycles: Lists the jar file's cyclic dependencies. - UnresolvedDependencies: Names of packages not found in the jar files in the directory being analyzed.