On 05/02/2008, at 6:35 AM, Christian Edward Gruber wrote:
The point is that neither futzing with modues, or hacking
classifiers is sufficient. But associative metadata might just be
the trick.
This is the approach we took with Eclipse Kepler as we mapped out the
repository. We actually
A classic use-case of this that I think is orthogonal to "javadoc" and
"sources" classifiers would be binary native artifacts per-platform.
Taking libc for a sec, (stupid example, but what the heck... You
might have:
libc-2.0.5-win32-win32.dll.
libc-2.0.5-openbsd-i386.so
libc-2.0.5-darwin
I can accept this, particularly if it leads to having dependency
metadata that is specific to these [formerly attached] artifacts.
Assemblies that contain their dependencies, when used as
dependencies, should affect dependency resolution differently than
the associated "naked" jar...which t
On 4-Feb-08, at 8:56 AM, John Casey wrote:
I'd tend to disagree about classifier not being a 'core' part of the
artifact system...it distinguishes a main artifact from one of its
derivatives, and serves as a pretty foundational part of how we
retrieve artifacts from existing remote reposit
I'd tend to disagree about classifier not being a 'core' part of the
artifact system...it distinguishes a main artifact from one of its
derivatives, and serves as a pretty foundational part of how we
retrieve artifacts from existing remote repositories. Without it, I
doubt that you can reco
