Copyright © 2009 Red Hat Middleware
Table of Contents
JBoss Tattletale is a tool that can help development teams getting an overview of the project they are working on or a product they depend on.
The tool generates reports that will show dependencies and general information that can help identify areas that needs attention such as minimizing the number of dependencies or eliminate duplicated class files from the class path.
JBoss Tattletale will help to improve the quality of your software project.
Jesper Pedersen acts as the lead for the JBoss Tattletale project. He can be reached at jesper (dot) pedersen (at) jboss (dot) org.
Have you ever found yourself frustrated with a ClassNotFoundException? Would you like to know what libraries are in your project and what they depend on? Would you like to get a full report on this stuff every time you run your ant build? If so you need to use the JBoss Tattletale project!
JBoss Tattletale is a tool that can help you get an overview of the project you are working on or a product that you depend on.
The tool will provide you with reports that can help you
Identify dependencies between JAR files
Find missing classes from the classpath
Spot if a class/package is located in multiple JAR files
Spot if the same JAR file is located in multiple locations
With a list of what each JAR file requires and provides
Verify the SerialVersionUID of a class
Find similar JAR files that have different version numbers
Find JAR files without a version number
Find unused JAR files
Identify sealed and signed JAR archives
Locate a class in a JAR file
Get the OSGi status of your project
JBoss Tattletale will recursive scan the directory pass as the argument for JAR files and then build the reports as HTML files.
JBoss Tattletale is licensed under GNU Lesser General Public License (LGPL) version 2.1 or later.
We hope that JBoss Tattletale will help you in your development tasks !
Please, visit the official JBoss Tattletale project page at http://www.jboss.org/tattletale/.
JBoss Tattletale can be downloaded in its binary form for easy installation.
The download location is: http://www.jboss.org/tattletale/downloads
Once downloaded extract the files by executing:
unzip jboss-tattletale-1.1.1.GA.zip
or
tar xzf jboss-tattletale-1.1.1.GA.tar.gz
depending on which archive type you downloaded.
JBoss Tattletale is now located in a folder under the directory you extracted the files into.
The JBoss Tattletale project is published in the JBoss Maven2 repository:
repository.jboss.org
under the group id of: org.jboss.tattletale
The artifacts for the project are
Developer snapshots are published on the JBoss Snapshots Maven2 server:
snapshots.jboss.org
If you want to experiment with the latest developments you may checkout the latest code from SVN. Be aware that the information provided in this manual might then not be accurate.
The anonymous SVN repository is located under:
svn co http://anonsvn.jboss.org/repos/tattletale/trunk/ tattletale-trunk
The developer SVN repository is located under:
svn co https://svn.jboss.org/repos/tattletale/trunk/ tattletale-trunk
The project is compiled using Java Development Kit 1.5 or higher, Apache Ant 1.7 or higher and Apache Ivy 2.1 or higher. Using
ant <target>
where target is one of
dist
Builds the distribution.
release
Builds the release archives.
doc
Builds the documentation for the project.
clean
Cleans the project of temporary files.
See the full list of targets in the main build.xml file.
The configuration of JBoss Tattletale is done through its
jboss-tattletale.properties
file.
The current configuration parameters includes:
Table 3.1. JBoss Tattletale configuration
| Key | Value |
|---|---|
| reports |
A comma separated list of which reports that should be generated.
The following reports are supported:
Default: All reports (reports=*) |
| classloader |
Specifies which classloader structure that should be used when scanning the archives.
Can be one of the following:
|
| profiles |
Specifies a comma separated list of profiles to resolve against.
The following profiles are supported:
|
| excludes |
A comma separated list of directories or files that should be excluded from the scan. F.ex.
**/server/**,myjar.jar |
| blacklisted |
A comma separated list of black listed classes or packages. F.ex.
com.mycompany.forinternaluseonly, com.partner.forinternaluseonly |
| scan |
A comma separated list of file extensions that should be scanned
Default: .jar |
| enableDot |
Should images be generated if the Graphviz DOT application is found
Default: true |
| graphvizDot |
The full path to the Graphviz DOT executable. This property is required if you want to generate
PNG files and the Graphviz bin directory is not on your path. F.ex.
graphvizDot=C:\\Graphviz2.26.3\\bin\\dot.exe orgraphvizDot=/opt/graphiz/bin/dot |
The load order for the configuration file is
NOTE: The classloader structure feature is currently based on directory structures and may therefore fail to identify archives that should be included in the reports. If you want to be sure that all archives are included use the NoopClassLoaderStructure plugin.
JBoss Tattletale supports filtering of the warnings and errors that the reports generates.
This functionality allows the user to filter out any warnings or errors that should be ignored and thereby allow the application to successful terminate.
The format of the filter properties file is
An example:
myreport=myfilter
The load order for the filter file is
See the individual reports for filtering support.
Running JBoss Tattletale is very easy
java -Xmx512m -jar tattletale.jar [-exclude=<excludes>] <sourcedir> [<outputdir>]
where the "sourcedir" is the directory that contains your Java archives and the optional "outputdir" parameter is the directory where you would like your reports to be generated. The "-exclude" option let you exclude directories or file on the command line - see the configuration file for syntax.
The main file will be generated under the output directory as index.html.
JBoss Tattletale will scan for Java Archives (.JAR) files.
JBoss Tattletale requires Java Runtime Environment 5 or higher.
JBoss Tattletale integrates with Apache Ant such that you can generate the reports directly from your build environment.
First, you need to add tattletale.jar, tattletale-ant.jar and javassist.jar to the Apache Ant classpath.
Second, you need to add the following to your project definition tag:
xmlns:tattletale="antlib:org.jboss.tattletale.ant"
That is it.
Alternative, you can do a taskdef for each task
<taskdef name="report"
classname="org.jboss.tattletale.ant.ReportTask"
classpathref="tattletale.lib.path.id"/>
See the Apache Ant documentation for additional instructions on installation.
Usage:
<tattletale:report source="${src.dir]" destination="${dest.dir}"/>
Table 4.1. Apache Ant: report
| Key | Value |
|---|---|
| source |
The directory that contains the Java archives. Multiple directories can be
scanned by separating each with the File.pathSeparator character - f.ex.
dir1:dir2 on Un*x.
Default: Current directory |
| destination |
The directory where the reports should be generated
Default: Current directory |
| configuration |
Path to the configuration file
Default: No value |
| filter |
Path to the filter file
Default: No value |
| reports |
A comma separated list of which reports that should be generated.
All reports can be selected by specifying "*". The following reports are supported:
Default: All reports |
| classloader |
Specifies which classloader structure that should be used when scanning the archives.
Can be one of the following:
Default: org.jboss.tattletale.reporting.classloader.NoopClassLoaderStructure |
| profiles |
Specifies a comma separated list of profiles to resolve against. All profiles can
be selected by specifying "*". The following profiles are supported:
Default: java5, java6 |
| excludes |
A comma separated list of directories or files that should be excluded from the scan. F.ex.
**/server/**,myjar.jar Default: Empty list |
| blacklisted |
A comma separated list of black listed classes or packages. F.ex.
com.mycompany.forinternaluseonly, com.partner.forinternaluseonly Default: Empty list |
| failOnInfo |
Fail the build if a failed INFO report is found
Default: false |
| failOnWarn |
Fail the build if a failed WARN report is found
Default: false |
| failOnError |
Fail the build if a failed ERROR report is found
Default: false |
| scan |
A comma separated list of file extensions that should be scanned
Default: .jar |
JBoss Tattletale integrates with Apache Maven such that you can generate the reports directly from your build environment.
To be able to use the Tattletale Maven plugin in your Maven project, you will have to add the following plugin declaration in the pom.xml of your project:
<build>
<plugins>
<plugin>
<groupId>org.jboss.tattletale</groupId>
<artifactId>tattletale-maven</artifactId>
<!-- The version of the plugin you want to use -->
<version>1.1.1.Final</version>
<executions>
<execution>
<goals>
<goal>report</goal>
</goals>
</execution>
</executions>
<configuration>
<!-- This is the location which will be scanned for generating tattletale reports
-->
<source>/absolutepath/to/source/dir</source>
<!-- This is where the reports will be generated -->
<destination>/absolute/path/to/reports/dir</destination>
</configuration>
</plugin>
</plugins>
</build>
Usage:
Once you have configured your project's pom.xml to include the tattletale-maven plugin, as explained earlier, you can generate the report by running the package goal on your project
mvn clean package
Table 5.1. Apache Maven: report
| Key | Value |
|---|---|
| source |
The directory that contains the Java archives. Multiple directories can be
scanned by separating each with the File.pathSeparator character - f.ex.
dir1:dir2 on Un*x.
Default: Current directory |
| destination |
The directory where the reports should be generated
Default: Current directory |
| configuration |
Path to the configuration file
Default: No value |
| filter |
Path to the filter file
Default: No value |
| reports |
Contains nested report elements of which reports that should be generated.
All reports can be selected by specifying "*". The following reports are supported:
Default: All reports |
| classloader |
Specifies which classloader structure that should be used when scanning the archives.
Can be one of the following:
Default: org.jboss.tattletale.reporting.classloader.NoopClassLoaderStructure |
| profiles |
Contains nested profile elements of profiles to resolve against. All profiles can
be selected by specifying "*". The following profiles are supported:
Default: <report>java5</report><report>java6</report> |
| excludes |
Contains nested exclude elements of directories or files that should be excluded from the scan. F.ex.
<exclude>**/server/**</exclude><exclude>myjar.jar</exclude> Default: Empty list |
| blacklisted |
Contains nested blacklist elements of black listed classes or packages. F.ex.
<blacklist>com.mycompany.forinternaluseonly</blacklist>
<blacklist>com.partner.forinternaluseonly</blacklist> Default: Empty list |
| failOnInfo |
Fail the build if a failed INFO report is found
Default: false |
| failOnWarn |
Fail the build if a failed WARN report is found
Default: false |
| failOnError |
Fail the build if a failed ERROR report is found
Default: false |
| scan |
A comma separated list of file extensions that should be scanned
Default: .jar |
The class dependants report will lists which classes depends on a specific class.
The class depends on report will lists which classes that a class depends on.
The dependants report will lists which archives depends on a specific archive.
The depends on report will lists which archives that an archive depends on.
Table 6.4. Depends On report
| Archive | Depends On |
|---|---|
| The archive | A list of archives which the archive depends on. Classes which can't be found are listed in italic |
Filter key is: dependson
Filter definition is:
archive=[class|package](,[class|package])*;
An example:
dependson=myjar1.jar=org.eclipse.*;myjar2.jar=com.mycompany.MyClass,com.mycompany.OtherClass
The graphical dependencies report will create GraphViz dot files that show the dependencies as graphics.
As an example you can generate a PNG image using
dot -Tpng myarchive.dot > myarchive.png
See the GraphViz documentation for a full description on how to generate these images.
The transitive dependants report will lists all archives depends on a specific archive.
The transitive depends on report will lists all archives that an archive depends on.
Table 6.7. Transitive Depends On report
| Archive | Depends On |
|---|---|
| The archive | A list of all archives which the archive depends on. Classes which can't be found are listed in italic |
Filter key is: transitivedependson
Filter definition is:
archive=[class|package](,[class|package])*;
An example:
transitivedependson=myjar.jar=com.mycompany.MyClass,com.mycompany.OtherClass
The circular dependency report will lists all archives that has a circular dependency with another archive.
Archives that are marked with "(*)" has the circular dependency. Note, that the circular dependency can be through a transitive dependency and not a direct dependency.
Table 6.8. Circular Dependency report
| Archive | Circular Dependencies |
|---|---|
| The archive | A list of all archives which the archive has a circular dependency on. |
Filter key is: circulardependency
Filter definition is:
[archive](,[archive])*;
An example:
circulardependency=myjar1.jar,myjar2.jar
The class location will lists which archives that contain a specific class file.
Filter key is: classlocation
Filter definition is:
[class|package](,[class|package])*;
An example:
classlocation=org.eclipse.*,com.mycompany.MyClass
The OSGi report will display the OSGi state of your project.
Table 6.10. OSGi report
| Archive | OSGi | Report | Manifest |
|---|---|---|---|
| The archive | The OSGi state of the archive | The OSGi report for the archive | A sample OSGi enabled MANIFEST file |
Filter key is: osgi
Filter definition is:
[archive](,[archive])*;
An example:
osgi=myjar1.jar,myjar2.jar
The sealed information report will display the sealed status of your project.
Table 6.11. Sealed information report
| Archive | Status |
|---|---|
| The archive | The status if the archive is sealed or unsealed |
Filter key is: sealed
Filter definition is:
[yes|on|true|no|off|false]
An example:
sealed=off
The signing information report will display the signing status of your project.
Table 6.12. Signing information report
| Archive | Status |
|---|---|
| The archive | The status if the archive is signed or unsigned |
Filter key is: sign
Filter definition is:
[yes|on|true|no|off|false]
An example:
sign=off
The eliminate jar files with different versions lists archives that have the same name but has a different version identifier.
Table 6.13. Eliminate Jar report
| Archive | Location |
|---|---|
| The archive | The list of locations that the archive is found |
Filter key is: eliminatejars
Filter definition is:
[archive](,[archive])*;
An example:
eliminatejars=myjar1.jar,myjar2.jar
The invalid version report lists archives that doesn't have a valid OSGi version identifier.
Table 6.14. Invalid version report
| Name | Location |
|---|---|
| The archive name | The location and version identifier for the archive |
Filter key is: invalidversion
Filter definition is:
[archive](,[archive])*;
An example:
invalidversion=myjar1.jar,myjar2.jar
The multiple jar files report will list classes that appear in multiple jar files.
Table 6.15. Multiple Jar files report
| Class | Jar files |
|---|---|
| The class | The list of archives where this class is found |
Filter key is: multiplejars
Filter definition is:
[package](,[package])*;
An example:
multiplejars=com.mycompany.mypackage1,com.mycompany.mypackage2
The multiple jar files fpr packages report will list packages that appear in multiple jar files.
Table 6.16. Multiple Jar files report (Package)
| Package | Jar files |
|---|---|
| The package name | The list of archives where this package is found |
Filter key is: multiplejarspackage
Filter definition is:
[package](,[package])*;
An example:
multiplejarspackage=com.mycompany.mypackage1,com.mycompany.mypackage2
The multiple locations report will list archives that appear in multiple locations under the scanned source directory.
Table 6.17. Multiple Locations report
| Name | Location |
|---|---|
| The archive name | The list of locations where the archive is found |
Filter key is: multiplelocations
Filter definition is:
[archive](,[archive])*;
An example:
multiplelocations=myjar1.jar,myjar2.jar
The Unused Jar report lists archives that isn't referenced from any other Jar archive in the distribution. This doesn't mean however that the archive isn't used since it could be referenced through Java Reflection or through metadata.
Filter key is: unusedjar
Filter definition is:
[archive](,[archive])*;
An example:
unusedjar=myjar1.jar,myjar2.jar
The black listed report will list archives that uses black listed APIs.
Table 6.19. Black listed report
| Archive | Usage |
|---|---|
| The archive name | The list of packages that uses black listed APIs |
Filter key is: blacklisted
Filter definition is:
[archive](,[archive])*;
An example:
blacklisted=myjar1.jar,myjar2.jar
The Java ARchive (JAR) report will provide you with an overview of the archive.
Table 6.21. No version report
| Key | Value |
|---|---|
| Name | The archive name |
| Class Version | The version identifier for the class files |
| Locations | The list of locations for the archive |
| Manifest | The manifest file |
| Signing information | The signing information for the archive |
| Requires | The list of required classes |
| Provides | The list of provided classes - including SerialVersionUID (if present) |
JBoss Tattletale generates its reports based on Java archives and not source code. Make sure that sourcedir you specify when running JBoss Tattletale contains the Java archives (f.ex. .JAR files) that you need scanned.
JBoss Tattletale needs to process the information it gathers in memory, so you need to provide enough memory for that to happen. You can ajust the -Xmx parameter of the command line below if you are using Sun's Java Runtime Environment.
java -Xmx1024m -jar jboss-tattletale.jar <sourcedir> [<outputdir>]
We can't cover every single issue in this guide, so feel free to drop by our forums to see if a solution has already been provided. Otherwise feel free to ask your question there.
Our forum is located at http://www.jboss.org/index.html?module=bb&op=viewforum&f=320