[MMTK] MMTK 2.2b1

Konrad Hinsen hinsen@cnrs-orleans.fr
Fri, 27 Apr 2001 20:12:59 +0200

> I got the (now current) beta 3 and managed to build it on Mac OS X 10.0.1 
> which required the following tweaks:
> 1) remove all refs to libm in setup.py (lib is subsumed in libc on OS X)

Sigh... Did they have to do this? Couldn't they at least provide an
empty libm to be compatible with thousands of Makefiles out there?

Anyway, I added a test for this, so it should work (from 2.2b4, next week).

> 2) setenv C_INCLUDE_PATH /usr/include/gcc/darwin/2.95.2/g++
>    to make cc find several #includes referenced by ncvalues.h, needed by
>    MMTK_trajectory.c

That's clearly not my responsability. As far as I can see, all the
includes in ncvalues are part of the C standard, so they should be on
some default include path. All I can do is to point out this problem
in the documentation.

> 3) in the same MMTK_trajectory.c, #include <limits.h> instead of <values.h>
>    which is deprecated on BSDs and absent on OS X
> 4) because of same, substitute MAXINT for INT_MAX in MMTK_trajectory.c,
>    of #define MAXINT INT_MAX

Those modifications should work everywhere, so I applied them to the

> I'm testing things right now, but my main peeve: the Amber forcefield seems 
> to be absent in this release!

Indeed! In fact I am having some problems with Distutils there.
No solution this week, I am afraid.

Release 2.2b4 should then work fine with Mac OS X. Thanks for testing!

Konrad Hinsen                            | E-Mail: hinsen@cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-
Rue Charles Sadron                       | Fax:  +33-
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais