General

What is this project all about?
This Maven plugin generates reports on issue found in the configured issue management system. Define reports and configure the query and columns if the issue information to be rendered in tables.

[top]


Why is this not part of the maven-changes-plugin?
The maven-changes-plugin is based on Java 1.4 and we did not want to pollute it with Java 1.5 code. Matthew Beermann patch adding a Mojo to access Bugzilla has been around for over a year, but nothing yet happened. In addition to that we also want to integrate some metrics reports which will take us further away from the intention of the maven-changes-plugin. We currently focus on Bugzilla. The maven-changes-plugin has a general approach to use different issue management systems. This would be too much work since we wanted to provide a practical solution that satisfies our needs quickly.

[top]

Issue Management Systems

Which issue management systems are supported?
Currently only Bugzilla is supported and tested with version 3.6. We use Mylyn to connect to Bugzilla, so other versions should work, too.



Prior to version 0.3.0 of the issues-maven-plugin version 3.2 of bugzilla was tested to work. We checked that version 2.22.7 is working with the earlier releases, but noticed that some identifiers of attributes supported by 3.2 are missing. So you may want to adjust the columns to be returned accordingly if you are using 2.22 (e.g. the assigned_to_name name used by our standard configuration is missing in 2.22).



Please refer to FAQ on attribute IDs to get help on determining the attributes supported by your current installation.

[top]


I don't know the attribute names of my issue management system! What can I do?
The issues plugin provides a property to collect all attribute names it encounters during the search. Use <logColumns>true</logColumns> in the configuration of the report set. To log the attribute IDs you can use for e.g. to specify column names.



The information is logged at INFO level and is introduced by the text Attribute IDs:.

[top]


I want to access an issue management system other than the primary system specified in the POM via issueManagement. What can I do?
You can override the default values for URL and system with issueManagementUrl and issueManagementSystem. Please note that currently only Bugzilla is supported so there is no gain in specifying the system's name.

[top]

Configuration

I define properties for different report sets, but it seems that the last declared report set sets the properties for all. What is wrong?
Probably you did not declare a unique identifier for each report set. Please check that it is provided as specified in the POM's documentation in the section reportSet.

[top]


For most of my reports I want a special component to be set. Therefore I put this component name in the configuration for all report sets. Unfortunatly one of the reports requires to override this value to not be limited to one component, but to query for all. Overriding the component property with the empty value does not work. What can I do?
There is a special value to select all components: _ALL_. This value can be used to override inherited component names. Otherwise simply do not specify the component property.

[top]


Is reporting on multi project supported?
Yes, it is. But configuration is not supported. It is easy to provide one report for each sub module and one large for the whole project. The task data is stored in the parent's project (according to inheritance) target/task-data folder. Please note that local task data is used without checking if it is out of date. Please refer to Caching Bugs for details.



Make sure that every submodule defines its own component and the parent module defines none. Please refer to Multi Projects on the usage page for an example.

[top]


Bug reports change, but the changes are not reflected in the reports generated. What is wrong?
For performance a bug once fetched is cached in-memory and written to the folder target/task-data if memory runs low. Whenever a bug is required the plugin checks if it is already in the cache. If the plugin finds the bug in target/task-data from a previous run, it uses this instead of fetching it from the remote issue management system.



If you want the bugs to be fetched from the remote system make sure that the folder target/task-data is cleared (e.g. by clearing the whole target folder with mvn clean.

[top]


Why is the component column rendered? I did not specify it with the columns property, but it always appears as the second column. What can I do to get rid of this column?
Per default, if some constraints are met (for details please refer to Adding Component Column) the component column is automatically included. Multi projects benefit from this feature. If the column should not be rendered please set includeComponentAtIndex to -1.

[top]

Installation

The dependency resolution fails with some eclipse artifacts telling me that the range does not match (or that eclipse artifacts are missing at all).

Couldn't find a version in [1.0.0-v20070606] to match range [1.0.0,2.0.0)

org.eclipse.equinox:app:jar:null

What went wrong?
There is a difference between the version definition expected by Maven and specified by eclipse. Fortunately to the maven-eclipse-plugin there is a way to handle this problem:

Type:

mvn eclipse:to-maven -DstripQualifier

and specify the installation directory of eclipse when requested on the command line. Version 3.4.0 should work since we depend on eclipse 3.4 and Mylyn 3.0.5.

You may want to download eclipse and select the "Eclipse IDE for Java Developers".

[top]