On Tue, Oct 18, 2011 at 5:28 AM, Fields, Christopher J <cjfie...@illinois.edu> wrote: > The BioPerl core developers (including myself) have decided to work on > breaking up the huge code base into separate distributions on CPAN, using > dependencies to install only the needed modules (something WAY overdue). I > noticed that several distributions have undergone similar paths (LWP being a > recent example). > > Any pointers we could use? Can this be gradually done (BioPerl is HUGE, > around 1000 modules), or should we have these all ready to go at once?
I've been pondering the same. For parts of the codebase that don't use much dynamic module loading it may be possible to automatize this using Perl::PrereqScanner. I think we could actually use some clustering algorithm after that (irony points for reusing biology knowledge there). Leon
