You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@maven.apache.org by Carlos Sanchez <ca...@gmail.com> on 2005/07/08 19:14:15 UTC

Re: [m2] Artifact in two different versions?

Ok, I can copy them in the repo

hibernate 2.x to groupId net.sf.hibernate
hibernate 3.x to groupId org.hibernate

ibatis 1.x to artifactId ibatis1
ibatis 2.x to artifactId ibatis2

do you think this is the right way?

On 7/8/05, Trygve Laugstøl <tr...@codehaus.org> wrote:
> On Fri, Jul 08, 2005 at 10:09:02AM +0200, Jörg Schaible wrote:
> > Hi Tryvge,
> >
> > Trygve Laugstøl wrote on Friday, July 08, 2005 10:02 AM:
> >
> > > On Fri, Jul 08, 2005 at 09:53:19AM +0200, Jörg Schaible wrote:
> > [snip]
> > >> As long as the libraries are not incompatible to themselves, it
> > >> should be possible to use both. This might not be the use case for an
> > >> application, but building a framework supporting those different
> > >> versions. We are in the same situation in the Nano camp.
> > >
> > > Maven has a strong concept of artifacts. A artifact is
> > > identified by the group id and the artifact id and it can have
> > > multiple versions.
> > >
> > > If a project has version 1 and version 2 artifacts that can
> > > exists in the same classpath at the same time they they are
> > > basically two different artifacts and should have different
> > > artifact ids. In the case of hibernate the artifacts should
> > > then be named something like this:
> > >
> > >   | Group id         | Artifact id  | Version
> > >   | org.sf.hibernate | hibernate2   | 2.0
> > >   | org.sf.hibernate | hibernate3   | 3.0
> >
> > Well, this is exactly what we did in M1 locally by introducing a
> > different artifactId and using the override mechanism to point to a
> > project relative version managed by the scm. But this is no longer
> > possible in M2 and I am looking for a solution. The problem is, that
> > there are quite some artifacts existing, where this situation applies.
> > Hibernate is just one example.
> 
> You can still do the same with Maven 2 by naming the artifacs differently
> and putting them in the repository. I really don't see the big difference.
> What we really should do is to get upstream to name their artifacts
> properly.
> 
> --
> Trygve
> 
> 
> BodyID:254544622.2.n.logpart (stored separately)
> 
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


Re: [m2] Artifact in two different versions?

Posted by Arik Kfir <ar...@gmail.com>.
+1


Carlos Sanchez wrote:

>hibernate should be in different groupIds because they have changed
>the package names.
>
>On 7/8/05, Trygve Laugstøl <tr...@codehaus.org> wrote:
>  
>
>>On Fri, Jul 08, 2005 at 10:14:15AM -0700, Carlos Sanchez wrote:
>>    
>>
>>>Ok, I can copy them in the repo
>>>
>>>hibernate 2.x to groupId net.sf.hibernate
>>>hibernate 3.x to groupId org.hibernate
>>>
>>>ibatis 1.x to artifactId ibatis1
>>>ibatis 2.x to artifactId ibatis2
>>>      
>>>
>>I don't think having different group ids is a good solution to this
>>problem and probably just confusing to the users. I'm honestly not sure
>>what the best solution is. Is it really possible to have both artifacts in
>>the classpath at the same time? At least I would assume that this would
>>depend on the ordering of the classpath.
>>
>>    
>>
>>>do you think this is the right way?
>>>      
>>>
>>Not sure, anybody else have a opinion?
>>
>>--
>>Trygve
>>
>>
>>BodyID:254779857.2.n.logpart (stored separately)
>>
>>
>>    
>>
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: users-unsubscribe@maven.apache.org
>For additional commands, e-mail: users-help@maven.apache.org
>
>
>  
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@maven.apache.org
For additional commands, e-mail: users-help@maven.apache.org


Re: [m2] Artifact in two different versions?

Posted by Trygve Laugstøl <tr...@codehaus.org>.
On Fri, Jul 08, 2005 at 10:27:51AM -0700, Carlos Sanchez wrote:
> hibernate should be in different groupIds because they have changed
> the package names.

Right, ok. Then you could actually have both in your class path at the
same time too. For Hibernate this actually makes sense now :) Still unsure
about Ibatis though.

--
Trygve

> 
> On 7/8/05, Trygve Laugstøl <tr...@codehaus.org> wrote:
> > On Fri, Jul 08, 2005 at 10:14:15AM -0700, Carlos Sanchez wrote:
> > > Ok, I can copy them in the repo
> > >
> > > hibernate 2.x to groupId net.sf.hibernate
> > > hibernate 3.x to groupId org.hibernate
> > >
> > > ibatis 1.x to artifactId ibatis1
> > > ibatis 2.x to artifactId ibatis2
> > 
> > I don't think having different group ids is a good solution to this
> > problem and probably just confusing to the users. I'm honestly not sure
> > what the best solution is. Is it really possible to have both artifacts in
> > the classpath at the same time? At least I would assume that this would
> > depend on the ordering of the classpath.
> > 
> > >
> > > do you think this is the right way?
> > 
> > Not sure, anybody else have a opinion?
> > 
> > --
> > Trygve

Re: [m2] Artifact in two different versions?

Posted by Carlos Sanchez <ca...@gmail.com>.
hibernate should be in different groupIds because they have changed
the package names.

On 7/8/05, Trygve Laugstøl <tr...@codehaus.org> wrote:
> On Fri, Jul 08, 2005 at 10:14:15AM -0700, Carlos Sanchez wrote:
> > Ok, I can copy them in the repo
> >
> > hibernate 2.x to groupId net.sf.hibernate
> > hibernate 3.x to groupId org.hibernate
> >
> > ibatis 1.x to artifactId ibatis1
> > ibatis 2.x to artifactId ibatis2
> 
> I don't think having different group ids is a good solution to this
> problem and probably just confusing to the users. I'm honestly not sure
> what the best solution is. Is it really possible to have both artifacts in
> the classpath at the same time? At least I would assume that this would
> depend on the ordering of the classpath.
> 
> >
> > do you think this is the right way?
> 
> Not sure, anybody else have a opinion?
> 
> --
> Trygve
> 
> 
> BodyID:254779857.2.n.logpart (stored separately)
> 
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


Re: [m2] Artifact in two different versions?

Posted by Carlos Sanchez <ca...@gmail.com>.
hibernate should be in different groupIds because they have changed
the package names.

On 7/8/05, Trygve Laugstøl <tr...@codehaus.org> wrote:
> On Fri, Jul 08, 2005 at 10:14:15AM -0700, Carlos Sanchez wrote:
> > Ok, I can copy them in the repo
> >
> > hibernate 2.x to groupId net.sf.hibernate
> > hibernate 3.x to groupId org.hibernate
> >
> > ibatis 1.x to artifactId ibatis1
> > ibatis 2.x to artifactId ibatis2
> 
> I don't think having different group ids is a good solution to this
> problem and probably just confusing to the users. I'm honestly not sure
> what the best solution is. Is it really possible to have both artifacts in
> the classpath at the same time? At least I would assume that this would
> depend on the ordering of the classpath.
> 
> >
> > do you think this is the right way?
> 
> Not sure, anybody else have a opinion?
> 
> --
> Trygve
> 
> 
> BodyID:254779857.2.n.logpart (stored separately)
> 
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@maven.apache.org
For additional commands, e-mail: users-help@maven.apache.org


Re: [m2] Artifact in two different versions?

Posted by Trygve Laugstøl <tr...@codehaus.org>.
On Fri, Jul 08, 2005 at 10:14:15AM -0700, Carlos Sanchez wrote:
> Ok, I can copy them in the repo
> 
> hibernate 2.x to groupId net.sf.hibernate
> hibernate 3.x to groupId org.hibernate
> 
> ibatis 1.x to artifactId ibatis1
> ibatis 2.x to artifactId ibatis2

I don't think having different group ids is a good solution to this
problem and probably just confusing to the users. I'm honestly not sure
what the best solution is. Is it really possible to have both artifacts in
the classpath at the same time? At least I would assume that this would
depend on the ordering of the classpath.

> 
> do you think this is the right way?

Not sure, anybody else have a opinion?

--
Trygve