On Tue, 2003-07-08 at 03:13, Vincent Massol wrote: > Hi, > > I'd like to reopen the historical data subject as this is something I > find very useful and I don't think I've seen answer to it. > > First, let me give you an example of what I'd like to achieve: Generate > a report which lists maven projects and which gives them a mark. The > mark would show the "best-behaving" projects. It would a function of the > activity and the number of time the project has broken the build. The > formula doesn't matter and can be improved (like adding the meantime > between build failure and build fixes, etc). > > What is important is that to achieve this we need historical data of > build results. And the question, as always is "where do we store build > results". > > Here's the solution I've thought about and for which I'd like to have > your feedback: > > - use a Maven build listener (as in Ant) which logs project builds and > failures in an XML format (BTW, that alone is needed to get a Gump-like > dashboard). An alternative is a preGoal/postGoal.
A build listener is definitely a good idea and is now possible with the refactored code. Not possible to do cleanly in the code that exists. Would be useful for all sorts of integration issues, for me specifically IDE integration. > - this build listener would store the result in the local repository, > under: Well, this could be specified. I don't think the behaviour should be baked into Maven as to what the listener does specifically. > What do you think? +1 on the listener I think the rest is outside the scope of Maven proper and what you talk of is really the basis of another project that focuses on continuous integration issues. > Thanks > -Vincent > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] -- jvz. Jason van Zyl [EMAIL PROTECTED] http://tambora.zenplex.org In short, man creates for himself a new religion of a rational and technical order to justify his work and to be justified in it. -- Jacques Ellul, The Technological Society --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
