[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [gemini-dev] Maven artifact IDs
|
On Tue, Aug 31, 2010 at 2:29 PM, Costin Leau
<cleau@xxxxxxxxxx> wrote:
Hi Les,
With regards to blueprint and symbolic name, if I recall correctly,
there were some Eclipse guidelines that I followed. The sym name needs
to be unique so using the domain/package names makes sense.
The bundle name on the other hand doesn't - it just needs to be a
friendly, human readable identifier.
Sure it could have been eclipse-gemini-blueprint-core but that to me,
seems to long and redundant hence the shorter version.
I was thinking it would be org.eclipse.gemini.blueprint.core. For example, SpringSource's commons-collections OSGi bundle has the following:
groupId: org.apache.commons
artifactId: com.springsource.org.apache.commons.collections
Which is quite long :)
I was hoping that at least some convention would take hold across the OSGi community, if for nothing more than being consistently easy to recognize during pom and/or ivy perusal. I do see your points though.
<aside>
I find SpringSource's naming for OSGi bundles a little confusing. To me, a Maven groupId should always be based on the 'group' or organization (and or project name, which is like a group within an organization) that is distributing the artifact, and the artifactId should reflect the actual api that is provided. For example, I think it'd make more sense to have the above revised to be:
groupId: com.springsource
artifactId: org.apache.commons.collections
</aside>
Anyway, thanks for the explanation - it is always appreciated!
Best,
Les
