As long as you're ok with downloading it (i.e. it's in the
repository), wouldn't the easiest thing to do be to depend directly on
CGLIB 1.x, but with <scope>provided</scope>?  It's even almost
semantically accurate, CGLIB 1.x functionality is provided by another
dependency: cglib-nodep-2.x

-Stephen

On 11/14/05, Brett Porter <[EMAIL PROTECTED]> wrote:
> This is a feature we have marked for 2.1 already. In the mean time,
> you need to do some surgery on your dependency tree, excluding the
> ones you don't want.
>
> - Brett
>
> On 11/14/05, Jörg Schaible <[EMAIL PROTECTED]> wrote:
> > Hello folks,
> >
> > first let me describe the scenario: CGLIB is used in a wide range of 
> > product as well as ASM. CGLIB itself is based on parts of ASM. CGLIB 1.x is 
> > dependent on ASM 1.x, but it provided also a cglib-full artifact, that 
> > contained the used sources from ASM 1.x. Therefore there are quite some 
> > projects (e.g. jmock-cglib), that depends on cglib-full.
> >
> > Major headache started, when ASM 2.x was released, since it is neither 
> > compatible to ASM 1.x nor you can use both ASM versions in the classpath. 
> > New CGLIB 2.x is still dependent on ASM 1.x, but they provide now a new 
> > artifact cglib-nodep 2.x, that contains again the necessary ASM 1.x 
> > classes, but they are located in a different package.  This allows a 
> > coexistence of CGLIB and the new ASM version in one classpath.
> >
> > The current problem with the transient dependencies are projects, with 
> > (transitive) dependencies on cglib-full-1.x and asm-2.x (like Groovy or 
> > some artifacts of NanoContainer). cglib-nodep-2.x is a compatible 
> > replacement for cglib-full-1.x, but I find no way to specify this anywhere.
> >
> > - Jörg
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > For additional commands, e-mail: [EMAIL PROTECTED]
> >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>


--
Stephen Duncan Jr
www.stephenduncanjr.com

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to