Re: Leakage from pluginManagement to plugins

2010-03-12 Thread Benson Margulies
Je pense que je ne mange pas des castor. On Fri, Mar 12, 2010 at 8:27 AM, Baptiste MATHUS wrote: > Yes, it's better. This way, it will take seconds for one to check if we're > having the same problem. > Just "design" your test project the best you can, like kind of a testcase. > So that the erro

Re: Leakage from pluginManagement to plugins

2010-03-12 Thread Baptiste MATHUS
Yes, it's better. This way, it will take seconds for one to check if we're having the same problem. Just "design" your test project the best you can, like kind of a testcase. So that the error just jumps out when maven is started. Describe the test project, what you wanted to do, what it does, and

Re: Leakage from pluginManagement to plugins

2010-03-12 Thread Benson Margulies
So, does anyone need an test case of this on a JIRA? On Fri, Mar 12, 2010 at 7:53 AM, Brian Fox wrote: > If the plugin is mentioned _at all_ even just group/artifact, then the > pluginManagement will kick in. This is how it works now. It should be > that you have to mention the executionId you w

Re: Leakage from pluginManagement to plugins

2010-03-12 Thread Brian Fox
If the plugin is mentioned _at all_ even just group/artifact, then the pluginManagement will kick in. This is how it works now. It should be that you have to mention the executionId you want to inherit instead of just assuming all of them. On Thu, Mar 11, 2010 at 10:37 PM, Benson Margulies wrote:

Re: Leakage from pluginManagement to plugins

2010-03-11 Thread Benson Margulies
Here, let me sketch this out... g a v ego g1 g a v super g2 If this is in the parent, and the child mentions g2 in it's //

Re: Leakage from pluginManagement to plugins

2010-03-11 Thread Brian Fox
If the execution is mentioned in the parent plugin section, it's inherited by the child, so yes it would run. If you don't want this, then the parent should have the configuration in pluginManagement (only) and then the child mentions the plugin in the plugins section. On Thu, Mar 11, 2010 at 1:12

Re: Leakage from pluginManagement to plugins

2010-03-11 Thread Anders Hammar
I think I need to see a project to tell if I think it's wrong or not. What I think you should start with is trying it with Maven 3.0-alpha-6 and see if it the same behavior. Many incorrect things in Maven 2 has been fixed in 3. /Anders On Thu, Mar 11, 2010 at 17:05, Benson Margulies wrote: > And

Re: Leakage from pluginManagement to plugins

2010-03-11 Thread Benson Margulies
Anders, If this sounds wrong to you, I'll put the effort into a test project. --benson On Thu, Mar 11, 2010 at 10:43 AM, Anders Hammar wrote: > Have you tested with Maven-3.0-alpha-6? > In either case, a test project to reproduce this would help. If you think > it's wrong, submit a jira with

Re: Leakage from pluginManagement to plugins

2010-03-11 Thread Anders Hammar
Have you tested with Maven-3.0-alpha-6? In either case, a test project to reproduce this would help. If you think it's wrong, submit a jira with that project attached. /Anderfs On Thu, Mar 11, 2010 at 16:36, Benson Margulies wrote: > As an additional hint, and possible explanation: > > I have a

Re: Leakage from pluginManagement to plugins

2010-03-11 Thread Benson Margulies
As an additional hint, and possible explanation: I have a project with multiple mojos in it. In the area where I'm having the problem, I've got one of them specified in build/plugins in the parent project, and the other specified in build/pluginManagement/plugins. I can't help wondering if there s

Leakage from pluginManagement to plugins

2010-03-11 Thread Benson Margulies
In the parent project, I declare a plugin in pluginManagement. In one of the children, I do \not/ reference it at all. Yet, help:effective-pom shows it in the build/plugins, and it's trying to run. Any suggestions?