> Actually I don't see that as a trick at all. This breaks my "principle of least surprise" test. Without being explicitly told what that pom fragment would do or running it myself, I wouldn't be entirely certain how Maven would be behave in the parent or in the children.
> If a future version of Maven were to break the feature (add the same > execution twice, once with inherited=true and the second with > inherited=false and the inherited=false configuration is closer so wins) > then I would consider that a serious bug. On that basis alone, I agree this could be perceived as "not a bug." But unless we are certain this is desired functionality -AND- there is a specific IT test to capture this functionality to avoid regressions, I'd be a little bit wary of using it myself. IMO the right way of doing this is to use pluginManagement and add 1-2 lines referring to this plugin in the various children poms where the plugin should be executed. This also satisfies the "least surprise" test. Wayne --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
