Vincent Massol wrote: >> 2) Problem of accessing plugin configuration data. See >> http://jira.codehaus.org/browse/MNG-2180 > > Kenney suggested to reparse the pom. I'll have a look at how it's done in > the Eclipse plugin. Right now, I have assumed that the location of the > clover databases is the same as the location defined in the parent.
That should work for now. It would be good to speculate on a better solution though for 2.1. You shouldn't have to read the config yourself, but I can't think of a good way to deal with a configured mojo for a subproject. I guess it could just be a list of configured mojos that you would have private get access to since you are in the same class? Seems ugly, still. I'd prefer those mojos to execute in some way and gather state to be dealt with at the end. > >> 3) Problem of having multiple reports in one plugin. In this case one >> report needs forking a custom lifecycle while the other must not fork a >> lifecycle. > > Still no idea how to handle this one. Seems we need the same executions > mechanism for report plugins that we have for build plugins. WDYT? We do. It's reportSets. The default is to execute *all* though, where in <build> the default is to run *none*. - Brett --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
