Re: Aether 0.9.0.M1

2012-10-02 Thread Jason van Zyl
Doesn't bother me if you use Aether directly. Even in a component that is ostensibly shared it is but one set of use cases. Everyone was arguing to shield the base artifact resolution service behind one facade, not the single set solution that is represented in the tree component. Extending the

preparations for release Maven Invoker

2012-10-02 Thread Robert Scholte
Hi, I'd like to prepare a release of the Maven Invoker somewhere next week. There are 3 issues left: https://jira.codehaus.org/browse/MSHARED-20 - Extend Invoker API to support the needs of the Release Plugin We'll need to move the code from org.apache.maven.shared.release.exec-package of th

Re: Aether 0.9.0.M1

2012-10-02 Thread Mark Struberg
I don't get it. Having shared/maven-dependency-tree hiding all that stuff will perfectly allow us to switch to another dependency resolution approach without having to change anything. Having maven plugins depend on aether directly is what creates tons of problems as those plugins would not work

Re: Aether 0.9.0.M1

2012-10-02 Thread Jason van Zyl
On Oct 2, 2012, at 10:15 AM, Hervé BOUTEMY wrote: > Le lundi 1 octobre 2012 19:34:27 Jason van Zyl a écrit : >> On Oct 1, 2012, at 4:38 PM, Hervé BOUTEMY wrote: >>> Given this milestone will be available in central repo [1], I don't see >>> any >>> problem: I created MNG-5354 to track this. >>

Re: Aether 0.9.0.M1

2012-10-02 Thread Hervé BOUTEMY
Le lundi 1 octobre 2012 19:34:27 Jason van Zyl a écrit : > On Oct 1, 2012, at 4:38 PM, Hervé BOUTEMY wrote: > > Given this milestone will be available in central repo [1], I don't see > > any > > problem: I created MNG-5354 to track this. > > Cool, thanks. > > > Of course, as requested by our de