www-repository mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Guillaume Nodet" <gno...@gmail.com>
Subject Re: trying to understand the xbeans dependencies
Date Tue, 03 Jul 2007 12:42:50 GMT
FWIW, xbean (@geronimo) is now publishing with org.apache.xbean
groupId only.  The org.xbean groupId was used before the migration
from Codehaus to Apache.
And I have no idea why the artifacts are in xbean groupId.  Note that the poms
are using org.xbean (see
http://repo1.maven.org/maven2/xbean/xbean-spring/2.1/xbean-spring-2.1.pom).

On 7/3/07, Steve Loughran <steve.loughran@gmail.com> wrote:
> the other graph I'm trying to understand is xbeans/xmlbeans. The code
> I'm ivyfying has an xmlbeans.jar in its lib dir that I am trying to
> match to artifacts (*)
>
> 1. There's an xbean/xbean/2.1 artifact that has a full m2 pom
> ftp://ibiblio.org/pub/packages/maven2/xbean/xbean/2.1/xbean-2.1.pom
>
> complete with seemingly excessive dependencies
>
> 2/ There's an xmlbeans/xbean which represents the apache xmlbeans
> product, which also has a 2.1 version
>
> 3. There's an org.xbean/xbean artifact that seems to share the pom of
> /xbean/xbean
> ftp://ibiblio.org/pub/packages/maven2/org/xbean/xbean/2.1/xbean-2.1.pom
>
> This is all a mess.
>
> 1. We shouldnt have both org.xbean and xbean, and if the latter must
> exist, surely its pom must be updated to accomodate the fact that it
> is, by virtue of being in a different place, a different artifact.
>
> 2. its messy for both Apache XMLBeans and Apache Geronimo xbeans to
> both be publishing xbeans JAR files. It means >1 apache project
> producing jars of the same name, which is going to lead to all kinds
> of confusion.
>
> I dont think we can fix (2) in the repository, but we ought to
> complain to the appropriate teams/pmcs that they are making things
> messy downstream.
>
> -steve
>
> (*) of course, I could just md5sum it and work backwards from that. If
> we had a form or URL that would do the mapping from an md5sum to
> artifact, it would be trivial
>


-- 
Cheers,
Guillaume Nodet
------------------------
Principal Engineer, IONA
Blog: http://gnodet.blogspot.com/

Mime
View raw message