If you had a repo manager proxying central this could be fixed in a minute
or two...

/Anders

On Thu, Sep 9, 2010 at 22:07, John Singleton <[email protected]> wrote:

> Unfortunately, our bcel dependency is a transitive dependency from a jBoss
> pom, so I can't easily change that one either.
>
> For the record, I do not disagree that strict pom validation is a good
> idea,
> nor do I disagree that it is the responsibility of the artifact producer to
> produce a correct pom.  I just wanted to point out that introducing
> mandatory validation without some sort of escape (e.g.
> -Dskip.pom.validation=true) will inconvenience some developers.  For now, I
> have to choose between hosting a few extra artifacts or remaining with
> maven
> 2.2.1 while pinging on a few developers to update their artifacts...
>
> John Singleton
>
> On Thu, Sep 9, 2010 at 2:14 PM, Wayne Fay <[email protected]> wrote:
>
> > > bcel/bcel/5.1 and ant-contrib/ant-contrib/1.0b3.  Both of these poms
> are
> > > structurally incorrect:
> >
> > bcel moved to o.a.bcel and has a 5.2 release with a proper pom file:
> >
> >
> http://mirrors.ibiblio.org/pub/mirrors/maven2/org/apache/bcel/bcel/5.2/bcel-5.2.pom
> >
> > ant-contrib 1.03b does look to be wrong, you should ping that
> > team/list to ask them to push a new build (1.03b is from Jan 2009)
> > that has, among other things, a proper pom file (and hopefully a more
> > sensible version number)
> >
> > > original question remains - is this maven behavior intended and, if so,
> > can
> > > we increase the urgency of fixing bad poms in maven central (
> > repo1.maven.org)
> > > so I don't have to host a lot of already-available artifacts just to
> fix
> > > their metadata?
> >
> > Fixing bad metadata is generally the responsibility of the dev team
> > who is producing the artifacts.
> >
> > Wayne
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [email protected]
> > For additional commands, e-mail: [email protected]
> >
> >
>

Reply via email to