[SciPy-user] Building Scipy from source

[Glen Shennan]

Thanks very much for that.  I was hoping to build everything from source
mainly for the learning experience.  I thought I got most of the way through
it but apparently not.  One day.  :)

2008/12/3 David Cournapeau <cournape at gmail.com>

> On Tue, Dec 2, 2008 at 9:34 PM, Glen Shennan <glen.shennan at gmail.com>
> wrote:
> > Hello,
> >
> > I'm new to Scipy (and Linux in general) and am trying to build Scipy from
> > source, following the directions on the official installation guide
> > beginning at the section titled "Building everything from source with
> > gfortran on Ubuntu".  I am running Ubuntu 8.04 (Debian, kernel version
> > 2.6.24-22) on a dual-core AMD 64 bit machine.  I can get through the
> > building of lapack, ATLAS, UMFPACK, FFTW without problems but I can't
> finish
> > off the numpy/scipy compile and was hoping someone here could enlighten
> me.
> > Numpy compiles but I'm not sure that it is working as intended/required
> and
> > the scipy build produces a huge string of errors ending with:
>
> Hi Glen,
>
> The easiest way to build both numpy and scipy is to avoid building
> atlas, blas and co by yourself. Those are difficult to build right.
>
> Do the following:
>
> sudo apt-get install g77 atlas3-base-dev atlas3-base
>
> Then remove the build directories in both numpy and scipy (to make
> sure you build from scratch), as well as the site.cfg files. You
> should then be able to build both numpy and scipy without any trouble.
> To avoid trouble with umfpack, you should try building scipy with the
> following command:
>
> UMFPACK=None python setup.py build
>
> Finally, although last numpy release is fine, you should build scipy
> from svn instead of 0.6.0. 0.6 is more than one year old; we are about
> to release 0.7, so the trunk should be fairly stable (and we would be
> able to help you better if there is any problem compared to 0.6).
>
> David
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.org
> http://projects.scipy.org/mailman/listinfo/scipy-user
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20081207/b1cbf042/attachment.html>