Hey guys, I know that I just joined on mailing list but I would like to share my experience with you. I am working on project which have more than 20 maven modules and we are managing just fine. We have nightly builds and deliveries each week.
In short: +1 for maven 0 for one module (project will explode in future if we have all in one component) Thanks, dinok On Wed, Oct 12, 2011 at 11:19 PM, Mattmann, Chris A (388J) < [email protected]> wrote: > Hey Enis, > > Honestly, I wasn't going to push it, but I've felt the same way since > pushing to roll 0.1.1 and actually getting Maven working. I'm super > comfortable with Maven and though I can fumble around and get > Ant + Ivy working it's certainly not my wheelhouse. > > My literal plan was to push forward on the GORA issue I filed to > roll forward the Maven poms from 0.1.1 (that mostly work), get them > fully working with 0.2-incubating, and then volunteer my time > to continue to maintain the poms. > > I'm +0 on collapsing into a single project -- Maven is really fine > with multi-module projects and it's quite a bit easier in my mind > to do that. I have no worries about maintaining the current module-based > structure for Gora so long as we're using Maven. > > Cheers, > Chris > > On Oct 12, 2011, at 11:07 AM, Enis Söztutar wrote: > > > Hi guys, > > > > I know we have discussed this previously in Maven vs Ant+Ivy flame wars, > and > > the consensus was staying with Ant+Ivy, and keeping Maven artifacts. > > However, looking at the recent list of issues, and dev threads, I seems > > unfortunate that because of the build system that we have right now, > builds, > > tests, nightlies and releases are becoming a major problem. And it is sad > > that most of the dev effort goes to the build system instead of improving > > core Gora. > > > > Initially, I had setup the project organization to be modular with and > > Ant+Ivy setup, which I guess, served well to this day. But since I cannot > > spend much effort for maintaining the build, and most of the active > > developers are more comfortable with Maven, I guess it is time to > simplify > > things a little and solve this build problem once and for all, so that we > > can spend more time focusing on core features. > > > > So what I am proposing is to have a volunteer for the build system who > wants > > to maintain the builds, and completely switch to Maven, and possibly > getting > > rid of the modular structure, and go with a one-module organization. > Having > > a modular project is great, and releasing different artifacts for hbase, > > cassandra, sql, tutorial, etc is the preferred way, but again, it may be > > causing more headaches. The volunteer maintainer will develop the > patch(es) > > for completely switching to Maven, and will oversee the build-related > > issues. > > > > What do you guys think? > > > > Thanks, > > Enis > > > ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > Chris Mattmann, Ph.D. > Senior Computer Scientist > NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA > Office: 171-266B, Mailstop: 171-246 > Email: [email protected] > WWW: http://sunset.usc.edu/~mattmann/ > ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > Adjunct Assistant Professor, Computer Science Department > University of Southern California, Los Angeles, CA 90089 USA > ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > >
