On 16 Dec 2006, at 06:53, Ralf W. Grosse-Kunstleve wrote:
We have not developed those two features yet. Fair enough about the smtbx… We can easily mimic the mmtbx,
I just wanted to say that you are most welcome to start the smtbx tree under the cctbx SF probject. This is just an offer, I'd be just as happy if you decide to start the tree somewhere else.
Oh, I see! Sorry, I misunderstood you then. Thanks for the offer. It would make sense indeed. Of course, our ultimate goal is to develop the equivalent of Phenix for small molecules, and that would not live on sourceforge.net/projects/cctbx either.
It is technically very easy to leverage the cctbx build system for other modules (I have at least 20 other modules floating around), i.e. so that you can say libtbx/configure.py smtbx and you'll have everything set up (PYTHONPATH, PATH, smtbx.python, etc.). The problem is it is not documented enough (just Michael Hohn's notes at http://cci.lbl.gov/~hohn/scitbx-tour.html). I hope to find time to write more detailed web pages over the holiday break.
I would highly appreciate that. By easy you probably mean it boils down to add a few instructions in a config file but my experience is that finding out those in the first place is not easy! Anyway, I am afraid I will come back to you with countless questions when I will start to develop with the cctbx. Thanks again, Luc