[MMTK] MMTK Windows 2000

Konrad Hinsen hinsen@cnrs-orleans.fr
Wed, 5 Mar 2003 22:14:58 +0100

On Wednesday 05 March 2003 15:40, Lawrence Kelly wrote:
> ATTN: John Michelsen or other Windows experts.

I am not a Windows expert, but I can confirm one thing:

> Tracing back I find that the problem is probably that
> Scientific_netcdf.pyd does not get compiled.

Definitely. MMTK_trajectory imports Scientific_netcdf, so if that module =
there, trajectory operations can't work. MMTK trajectories are netCDF fil=

> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
> Building 'Scientific_netcdf' extension
> skipping Src/Scientific_netcdf.c
> (build\temp.win32-2.2\Release\scientific_netcdf.o up-to-date)
> writing build\temp.win32-2.2\Release\Scientific_netcdf.def
> e:\mingw\bin\dllwrap.exe -mno-cygwin -mdll -static --entry _DllMain@12
> --output-lib build\temp.win32-2.2\Release\libScientific_netcd
> f.a --def build\temp.win32-2.2\Release\Scientific_netcdf.def -s
> build\temp.win32-2.2\Release\scientific_netcdf.o -Lg:\Python22\netcd
> f-3.5.0\lib -Lg:\python22\libs -lnetcdf -lpython22 -o
> build\lib.win32-2.2\Scientific\win32\Scientific_netcdf.pyd
> build\temp.win32-2.2\Release\scientific_netcdf.o(.text+0x2e2):Scientifi=
>tcdf.c: undefined reference to `ncerr'

Just guessing, but could this be a problem due to a reference to a global=
variable in a shared library? In that case, it might help to use the one-=
distribution of netCDF and link it together with Scientific_netcdf, inste=
of relying on the shared library. I used this (for different reasons) in =
Sharp Zaurus port of Scientific_netcdf, so I can confirm that it works ve=

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