Jason van Zyl wrote:
I think this is a regression:
http://jira.codehaus.org/browse/MNG-3922
Attached artifacts with classifiers seem to have gotten hosed. Works in
2.0.8. No worky in 2.0.9.
I don't get that working with Maven 2.0.8 either, i.e. I don't believe
it's a regression, so could be postponed.
The important point in this issue is the order of plugin executions.
Maybe a change in parent inheritance or profile merging between 2.0.8
and 2.0.9 produced an order that worked for Max, hard to says more
without his build logs.
While it looks unintentional to me that an executionProject forked by
one mojo makes it into the execution of another mojo, it looks even more
strange that the Shade Plugin attaches the artifacts to the
executionProject when existent, considering that the shade mojo itself
does not fork the lifecycle. The code in question was introduced in [0]
but I can't imagine a use case where this would be required. So, should
we also fill a bug for the Shade Plugin and remove any references to the
executionProject?
Benjamin
[0] http://fisheye.codehaus.org/changelog/mojo/?cs=3821
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org
