incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jason van Zyl <ja...@maven.org>
Subject Re: [policy] incubating projects and maven repositories v1.0
Date Mon, 28 Aug 2006 00:58:34 GMT

On 27 Aug 06, at 6:28 PM 27 Aug 06, Craig L Russell wrote:

>
> On Aug 27, 2006, at 2:22 PM, Jukka Zitting wrote:
>
>> Hi,
>>
>> On 8/27/06, Jason van Zyl <jason@maven.org> wrote:
>>> There was discussion that incubator repository would not be  
>>> sync'd to
>>> the central repository but I don't really see much point in this.
>>> [...]
>>> Also someone may make an repository request to place an incubator
>>> artifact in the central repository and at this point a policy
>>> mandated here would conflict with someone's right to redistribute
>>> artifacts created in the incubator. I don't really want to get into
>>> the business of policing repository requests. I think it is in the
>>> best interests of the  incubating projects to have the incubator
>>> repository sync'd to Maven's central repository. The source of
>>> artifacts for incubating projects is clear from the version so I
>>> don't think there will be any confusion by consumers of these
>>> artifacts and as such I don't really see any downside to allowing  
>>> the
>>> sync to Maven's central repository.
>>
>> Me neither. But why do we then need the separate incubator repository
>> if the artifacts get synchronized with the central repository?
>>
>> As I mentioned in the thread before the Incubator Maven repository  
>> was
>> created, it makes more sense to enforce an "incubator" label on the
>> artifact versions than enforcing a specific "incubator" repository.
>> Especially since there is no way for us to really enforce that
>> repository policy.
>
> I agree. I understand that we want users who choose to use an  
> incubating project's artifacts to declaratively state that. If the  
> artifact's name contains "incubating" then it's pretty clear.
>
> The only thing that muddles things for me is when using an m2  
> repository that contains a non-incubating project with a dependency  
> on an incubating project. Then, a project that depends on the  
> project that is itself not in the incubator might not even know  
> that it's using an incubating project.

The standard dependency report shows all transitive dependencies so  
you can definitely see what you're using:

http://maven.apache.org/continuum/dependencies.html

Could be a little better in display but you can see what's in your  
dependency set. There are also things like the netbeans m2  
integration which draws nice graphs of the dependencies.

>
> Can this happen? Is it likely?

If you don't look at the dependency report or a graph then I suppose  
it could.

> Is there any policy that we can put into place to avoid that  
> projects with dependencies on projects with incubating projects  
> accidentally have this dependency?
>
> Perhaps Maven can help by not allowing the transitive dependency on  
> incubating projects? Just thinking out loud...
>

If you can clearly see what you have in a report so I'm not sure we  
would want to put any special rules in place to deal with artifacts  
from incubator.

> Craig
>
>
>>
>>
>> BR,
>>
>> Jukka Zitting
>>
>> -- 
>> Yukatan - http://yukatan.fi/ - info@yukatan.fi
>> Software craftsmanship, JCR consulting, and Java development
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
>> For additional commands, e-mail: general-help@incubator.apache.org
>>
>
> Craig Russell
> Architect, Sun Java Enterprise System http://java.sun.com/products/jdo
> 408 276-5638 mailto:Craig.Russell@sun.com
> P.S. A good JDO? O, Gasp!
>

Jason van Zyl
jason@maven.org




---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Mime
View raw message