[SciPy-user] Compiling extensions in a virtual environment [fixed]

Pierre GM pgmdevlist@gmail....
Tue Mar 17 09:54:18 CDT 2009


OK,
virtualenv magic must have hexed me somehow. After updating to the  
latest SVN in *and* outside of my virtual environment, the problem  
seems to be fixed.
If any regular user of virtualenv could enlighten us and tell us what  
files block the other, that'd be great.
Thx again.
P.


On Mar 17, 2009, at 4:11 AM, David Cournapeau wrote:

> On Tue, Mar 17, 2009 at 4:00 PM, Pierre GM <pgmdevlist@gmail.com>  
> wrote:
>> All,
>> I'm trying to compile scikits.timeseries in a virtual environment
>> (created w/ virtualenvwrapper), and I'm running into a puzzling  
>> problem.
>>
>> When I create a new virtual environment, install numpy in it (from
>> SVN, r6685), I'm able to compile the scikit without problem. However,
>> after removing the build and switching to another existing virtual
>> environment, I get a:
>> error: build/src.macosx-10.5-i386-2.5/gfortran_vs2003_hack.c: No such
>> file or directory
>> Where does this error come from ? As the virtual environment that
>> causes problem is the one that I work with all the time, I'd like to
>> know what's happening.
>
> I am sorry, I am the one to blame for this. I removed the offending
> code - you should make sure that you have updated numpy correctly (the
> revision 6685 has already the buggy code removed). As it introduced a
> bug in numpy.distutils, you should make sure which numpy you are using
> to build other extensions. This is a numpy problem at root, but I
> can't really help you for virtualenv - they do make some import magic
> which may make the problem more difficult to track down.
>
> cheers,
>
> David
> _______________________________________________
> SciPy-user mailing list
> SciPy-user@scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-user



More information about the SciPy-user mailing list