Hi folks. Firstly, thanks to the ivy team - ivy is exactly what I was looking for. Especially the transitive dependencies - it's perfect for describing multiple use cases.
I am wondering what your plans are for something like jayasoft's old ivyrep repository? I would love to have somewhere to read (and publish) ivy files, both for pre-existing open source libraries and my own projects. To provide some background, I've been using ivy for about a year now, having successfully migrated a large codebase from a monolithic ant build, via a multi-module maven2 build (painful and verbose) and then ant-driving-maven (just verbose). We decided to roll our own ivy.xml files for each external dependency, because the maven poms were trying to pull in every possible dependency. For most of the jars we used there were no pull-through dependencies we cared about so we used a boilerplate empty ivy file. For the others (such as some of the spring jars) it was pretty simple to put together the respective dependencies and have them pulled through transitively. We ended up with our own "local" ivy repository that just had the ivy files, and the jars themselves would be pulled down and cached from the maven repo at ibiblio. What I would like is a public version of our local repository, which it looks like ivyrep used to be before ivy moved to the apache incubator. Thanks again for a great tool. The post-resolve tasks are a stroke of genius. Cheers, Dan
