Once upon a time, Michael Heuer wrote: > > [EMAIL PROTECTED] mailing list (which is, after all, > > intended for people who are participating in, or at least > > actively following, development). If there's anyone > > who strongly considers this to be silly/evil/spammy, > > please speak up soon. > > > > Coming soon: > > > > - Similar reports for the BioJava 1.3x maintainance branch > > (I'm currently trying to work out the least-nasty CVS hack > > to do this). > > > > - Integration with nightly build and test reports. > > Using maven for this?
Not at the moment. When I last looked at Maven, I was a bit worried that it didn't seem to make much distinction between the useful functionality is was offering and the policy decisions it was trying to enforce. I'd be happy to try it out with a new project, but I don't want to roll over an existing BioJava-sized project without a lot of testing and consultation first. Maybe we should try Maven-izing a few BioJava subprojects (biojava-ensembl might be a reasonable candidate) to see how it works. > An additional useful report would be an unit test coverage report. Clover > is a pretty good tool for this, and can be integrated with maven (and ant, > of course). > > > http://www.thecortex.net/clover > > It requires a license, but the O|B|F could request a free license for use > in the biojava project. Have you looked at all at Quilt: http://quilt.sourceforge.net/ This still seems to be a bit of a work-in-progress, but it sounds like the basics are working, and it's Apache licenced. I'm tempted to give this a try, anyway. By the way, nightly builds are now going into: http://www.derkholm.net/autobuild/ I need to do a little more work on the reporting side, but this should be going public in the next few days. Thomas. _______________________________________________ Biojava-l mailing list - [EMAIL PROTECTED] http://biojava.org/mailman/listinfo/biojava-l
