you can indeed help me a bit, but with general guidance. I'm a maven noob and
therefor a bit lost. Here are my thoughts on this, please let me know if you
don't know what i'm speaking about:
Other hudson plugins (i looked into the violations plugin
http://wiki.jenkins-ci.org/display/JENKINS/Violations) seem to integrate into the
different maven goals to check if they are like "pmd:pmd" or
"checkstyle:checkstyle". They collect the report data directly afterwards for
these maven plugins.
For crap4j, there is the maven-crap4j-plugin
(http://jmd.sourceforge.net/maven-crap4j-plugin/index.html). It collects crap4j
data in the "crap4j:crap4j" goal, but only as part of the "site" phase.
As far as i can see, it would be sufficient to gather the report data at the end
of the build, from a well-known location, in the case of this maven plugin it
would be $WORKSPACE/target/site/crap4j-reports/report.xml
Am i missing something important? My test setup so far is to fetch a maven2 open
source project into my hudson (in my case, its the dom4j project
(http://www.dom4j.org/)), have the crap4j maven plugin included in the pom.xml
and run "clean install site" for each build.
Sorry if this sounds weird. Perhaps i have to ask other plugin authors about
this. I'm not too keen with maven to have the complete overview.