Closed bruceravel closed 10 years ago
Very cool! I'm up to my eyeballs in detector and data collection code and trying to something working for the start of the fall run. But when I get a chance, I'll be very excited to give this a spin....
On Thu, Oct 2, 2014 at 3:14 PM, Bruce Ravel notifications@github.com wrote:
Merged #18 https://github.com/xraypy/feff85exafs/pull/18.
— Reply to this email directly or view it on GitHub https://github.com/xraypy/feff85exafs/pull/18#event-173480204.
--Matt Newville
I made it! I now have a comprehensive and useful tool for generating individual paths given a scattering geometry. There is a Fortran entry point (libonepath.a) and a C wrapper around that (libfeffpath.c). As a proof of principle, I made a fairly simple Perl wrapper using SWIG.
I have also improved upon the build system so that it now installs to a sensible location, albeit only on a linux machine. The build scripts will need work for Windows or the Mac.
For now, cd-ing to the
src
folder and doingwill compile everything, install the various stand-alone "modules" to
/usr/local/bin
, and install various static libraries to/usr/local/lib
. The static libraries includelibonepath,a
andlibfeffpath.a
as well as libraries for all the various parts of feff (with names likelibfeffmath.a
andlibfeffatom.a
to avoid name collisions with other packages).If SWIG is installed, the
scons
step will generate the various SWIG files for the perl wrapper. Then thescons install
step will copy them into thewrappers/perl
folder so that the Perl wrapper can be built using Perl's normal build tools.The next steps are to write good documentation for everything and to incorporate individual paths into the overall unit testing framework.