[SciPy-User] Does scipy binary install libgfortran.dylib?
Mon Feb 27 06:28:30 CST 2012
Excellent, so it can be done.
The next question is: How?
I found a suggestion from Brian Toby on the web from this summer, but that
didn't work form me. This is what I did:
On my Mac Terminal I type:
LDFLAGS='-undefined dynamic_lookup -bundle -static-libgfortran -static-libgcc'
f2py -c -m besselaes besselaes.f95
This nicely creates the extension, which I can run on the machine I
created it on, but if I move it to a
machine that doesn't have the libgfortran.3.dylib file, it doesn't run
and complains about that, so I
conclude that the dynamic link failed (the size of the extension
doesn't change when I set the LDFLAGS,
which I thought was a bad omen). Any thoughts? Am I doing something wrong?
Thanks for any help,
Date: Sun, 26 Feb 2012 17:38:19 +0100
> From: Ralf Gommers <email@example.com>
> Subject: Re: [SciPy-User] Does scipy binary install libgfortran.dylib?
> To: SciPy Users List <firstname.lastname@example.org>
> Content-Type: text/plain; charset="iso-8859-1"
> On Sun, Feb 26, 2012 at 11:49 AM, Mark Bakker <email@example.com> wrote:
> > Hello List,
> > Does Scipy install the correct version of libgfortran.dylib? Does it
> > simply put it in /usr/local/lib/ ?
> > I am trying to distribute my own package which includes FORTRAN
> > and when installing on a brand new machine it complains that
> > libgfortran.3.dylib cannot be found. I was wondering how Scipy handles
> > (and thanks to Ralf Gommers for helping me so far, but I haven't been
> > to solve this).
> Sorry for not being clearer before - I only knew that this worked for
> scipy, not exactly how it worked. After some digging I found this in the
> 0.7.1 release notes: "Mac OS X binary installer is now a proper universal
> build, and does not depend on gfortran anymore (libgfortran is statically
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the SciPy-User