Xavier, Your logic does make sense to me and I realize that this is not an easy decision, as you pointed out in a later e-mail. My solution currently is to use IvyRep as the first repository and Ibiblio as the second in a repository chain. This allows me to get ivy.xml files whenever they're available. Though, I wasn't aware of the IvyRep Sandbox, so that may be another repository that i would throw in the chain. I realize that this is a complicated setup, but maybe something like this makes sense as the default? My thought here is that if you have the Ivy defaults set to use Ibiblio exclusively, you run the risk of the Ivy community not having enough reason to grow IvyRep or change Ibiblio to contain ivy.xml files. This would make Ibiblio and pom the long term solution, which doesn't seem appropriate. Regarding your notion of distributed repositories, I can see that working. The only concern is that every time you add a dependency, you most likely have to add another include in your ivyconf. This is a little more work than just having one central repository. Though, the idea of every project owning their own repository sounds much more robust and "cleaner" than one central repository containing every module in existence. Scott
Xavier Hanin <[EMAIL PROTECTED]> wrote: On 3/15/07, Scott Goldstein wrote: > > I'm not currently a contributor for this project, but I hope you don't > mind me asking a question on this topic because I have a interest in Ivy. Anybody is welcome to give an opinion! I agree that maintaining two repositories does not make sense. However, I > know that the ivy repository has ivy.xml files for each module. Is the > plan to add these to ibiblio? If not, what information/functionality would > be lost? If none currently, it certainly seems that some flexiblity would > at least be lost, as there will be a dependency on the pom format. The problem is that if you look at ivyrep, you will see that there are only a few modules for which we have ivy.xml files. On the other hand, in the ivyrep sandbox you've got much more, and I think most people either use ivyrep sandbox directly, or simply download ivy files from the sandbox and adapt them to their own needs. So this won't be lost, the sandbox (and even ivyrep) will still be available. The main change is that it won't be the default settings anymore, and that you will have to expressly depend on them if you want. For the dependency on the pom format, Ivy already support it (with bugs in 1.4.1, but the head version is getting more and more stable on this subject), so it's not really a problem to have this dependency. The idea is then to leverage the existing maven 2 repository, which has the advantage to have metadata for a lot of modules (even if their quality is not always good), and then use Ivy file only when necessary (for your own modules, for third party modules for which you need the Ivy flexibility, ...). Then you get the best of both worlds, and with Ivy this integration can be done seamlessly. Then if we see that users would be interested to share their Ivy files in a central repository, we will talk with the repository team to see if they are ok to host Ivy files. But this require an important amount of work of maintenance, and the Ivy community is not strong enough for the moment IMHO. Another point is that with Ivy you can quite easily have distributed repositories, and I'm not sure that a central repository is the best. If Ivy gain momentum, I would prefer seeing each project using Ivy provide their own Ivy repository with their modules, and an ivyconf.xml for this repository with proper per module configuration of resolvers. Then using their repository would be as simple as including their ivyconf.xml in yours (using the include feature). The big advantage is that then each project is responsible for the maintenance of their own repository, and as a user you can only include repositories in which you are interested. Then our work in the Ivy team would only be to reference the repositories, instead of maintaining a huge set of metadata. But maybe I'm dreaming :-) Does it make sense for you? Do you see use cases where we should consider things differently? - Xavier Scott > > Gilles Scokart wrote: > +1, I don't see the purpose of maintaining ivyrep when we have ibiblio > pom, > and if it is not maintained, it should not be used. > > Gilles > > > -----Original Message----- > > From: Xavier Hanin [mailto:[EMAIL PROTECTED] > > Sent: mercredi 14 mars 2007 9:38 > > To: [email protected] > > Subject: Ivyrep > > > > Hi, > > > > Ivy is still packaged with the ivyrep resolver, which is even used by > the > > default settings. The problem is that the official ivy repository is no > > more > > maintained, and I think that people starting using Ivy now would prefer > > using the maven2 repository with their poms, instead of what can be > found > > in > > the small ivyrep. Another problem is conflictual namespaces, because > > maven2 > > and ivyrep do not share the same naming conventions, it makes it harder > to > > use the two together (even if Ivy namespace feature makes it possible). > > > > So I think we should better use a maven 2 resolver (i.e. ibiblio with > > m2compatible set to true) in the default settings as the default public > > resolver, and deprecate the use of ivyrep without specifying the root. > > > > What do you think? > > > > - Xavier > > >
