[MMTK] numpy problem?

Konrad Hinsen research at khinsen.fastmail.net
Tue Nov 2 07:42:20 UTC 2010


On 1 Nov 2010, at 21:35, Ramgopal Mettu wrote:

> building 'Scientific_affinitypropagation' extension
> gcc-4.2 -fno-strict-aliasing -fno-common -dynamic -DNDEBUG -g - 
> fwrapv -Os -Wall -Wstrict-prototypes -DENABLE_DTRACE -arch i386 - 
> arch ppc -arch x86_64 -pipe -IInclude -I/System/Library/Frameworks/ 
> Python.framework/Versions/2.6/lib/python2.6/site-packages/numpy/core/ 
> include -I/System/Library/Frameworks/Python.framework/Versions/2.6/ 
> include/python2.6 -c Src/Scientific_affinitypropagation.c -o build/ 
> temp.macosx-10.6-universal-2.6/Src/Scientific_affinitypropagation.o - 
> DNUMPY=1
> In file included from Src/Scientific_affinitypropagation.c:14:
> Include/Scientific/arrayobject.h:2:30: error: numpy/oldnumeric.h: No  
> such file or directory
>
> This latter error in turn leads to a bunch of other errors and I  
> cannot finish building Scientific. Now, I have version 1.2.1 of  
> numpy. Is this compatible with Scientific? Do I need a newer  
> version, or am I missing something else? Thanks!

In principle NumPy 1.2 should be fine, but I haven't tested anything  
with it for a long time.

Given the paths printed above, there should be a file

/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/ 
site-packages/numpy/core/include/numpy/oldnumeric.h

on your computer but apparently it isn't because the compiler cannot  
find it. If the file isn't there, your NumPy installation is  
incomplete or broken for whatever reason. If it is there, well, then  
there's something wrong with gcc - but I haven't seen that case yet!

As a side remark, it looks like you are on a Mac (MacOS 10.6) and use  
Apple's Python interpreter. Apple advises against installing any add- 
on modules into their Python interpreter, both to protect their own  
Python scripts against errors introduced by other packages and to  
protect your Python packages against changes during system updates. In  
the long run, you will be better off installing a second Python  
interpreter for use with NumPy, MMTK, etc.

Konrad.
--
---------------------------------------------------------------------
Konrad Hinsen
Centre de Biophysique Moléculaire, CNRS Orléans
Synchrotron Soleil - Division Expériences
Saint Aubin - BP 48
91192 Gif sur Yvette Cedex, France
Tel. +33-1 69 35 97 15
E-Mail: research at  khinsen dot fastmail dot net
---------------------------------------------------------------------




More information about the mmtk mailing list