[Numpy-discussion] has f2py changed from 1.0.4 to 1.1.1 ????

mark markbak@gmail....
Wed Aug 13 03:15:40 CDT 2008

Charles and Robert -

I am running Python 2.5.2 on Windows XP.

I use the Enthought distribution, which comes with numpy 1.0.4. f2py
works fine using mingw32 and gfortran. My command line is:

python c:\python25\scripts\f2py.py -c -m besselaes --compiler=mingw32
--fcompiler=gnu95 besselaes.f90

Next I upgraded to numpy 1.1.1 (most importantly, because matplotlib
0.98 requires numpy 1.1.x). I run exactly the same command line and
get the error:

Ignoring "Python was built with Visual Studio 2003;
extensions must be built with a compiler than can generate compatible
Visual Studio 2003 was not found on this system. If you have Cygwin
you can try compiling with MingW32, by passing "-c mingw32" to
setup.py." (one should fix me in fcompiler/compaq.py)
running build
running scons
No module named msvccompiler in numpy.distutils; trying from distutils

I don't understand, because distutils still contains the files
msvccompiler.py and .pyc.

I removed numpy 1.1.1 and reinstalled the numpy1.0.4 egg, but that
didn't help. I had to reinstall the entire Enthought distribution to
get f2py to work again.

Any help is greatly appreciated.


On Aug 13, 4:25 am, "Robert Kern" <robert.k...@gmail.com> wrote:
> On Tue, Aug 12, 2008 at 20:12, Charles R Harris
> <charlesr.har...@gmail.com> wrote:
> > On Tue, Aug 12, 2008 at 3:21 PM, mark <mark...@gmail.com> wrote:
> >> Hello all -
> >> I have been banging my head against the wall.
> >> f2py worked fine on windows under 1.0.4 (Python 2.5)
> >> After upgrading to numpy 1.1.1, f2py keeps complaining it cannot find
> >> msvccompiler, yet it is still in the same place (under distutils) as
> >> it was under 1.0.4.
> >> Did something change? I read something about some f2py branch that was
> >> ended in 1.1.1.
> >> Any suggestions on how to solve this?
> > 1.1.1 should be the  same as the trunk f2py wise, so we need to figure this
> > out. Robert might have an idea as to what is going on.
> Without more information (like a traceback, the Fortran compiler that
> the OP is trying to use, and the --fcompiler, etc. options used), not
> really.
> --
> Robert Kern
> "I have come to believe that the whole world is an enigma, a harmless
> enigma that is made terrible by our own mad attempt to interpret it as
> though it had an underlying truth."
>  -- Umberto Eco
> _______________________________________________
> Numpy-discussion mailing list
> Numpy-discuss...@scipy.orghttp://projects.scipy.org/mailman/listinfo/numpy-discussion

More information about the Numpy-discussion mailing list