[IPOL discuss] GSL for IPOL [Re: OpenCV and IPOL]

Tristan tristan.dagobert at cmla.ens-cachan.fr
Mon Aug 26 20:00:12 CEST 2013


Hello.

The GSL team has recently released the 1.16 version. According to some 
of developers' discussion on its forum,
it seems that this version is be more compatible with the last microsoft 
compiler Visual Studio :
http://lists.gnu.org/archive/html/bug-gsl/2013-07/msg00030.html

I have not tested its compilation yet on MSCV but conversion scripts are 
available here  :
http://gladman.plushost.co.uk/oldsite/computing/gnu_scientific_library.php

I transmit you the detailed procedure of the 1.15 compilation. 
Principles of 1.16 compilation are similar.


Le 26/08/2013 18:03, Juan Cardelino a écrit :
>
> I could help with te windows version.  ve compiled  italia a couple of 
> times .
>
> El 26/08/2013 12:58, "Nicolas Limare" 
> <nicolas.limare at cmla.ens-cachan.fr 
> <mailto:nicolas.limare at cmla.ens-cachan.fr>> escribió:
>
>     Hi,
>
>     > in my opinion we should allow GSL [1], since it implements many of
>     > the operations needed in image processing or CV programs
>     > (eigensystems, minimization, wavelets, linear algebra, least-squares
>     > fitting, splines, etc).
>
>     I agree, GSL is useful and very stable. The only question was if it
>     would be usable for people working on Windows. GSL is not written with
>     windows compatibility in mind and the authors do not guarantee it is
>     usable on this platform. There are many blogs and websites explaining
>     how the compilation of GSL on Windows is possible, but they are always
>     specific to a given version of GSL and I was wondering if we could
>     rely on this information for the long term.
>
>     Tristan did a detailed study of this matter, and if I remember
>     correctly the conclusion was:
>     * The standard C compiler on Windows, MSVC, can not compile the
>       official version of the GSL because of non-standard behaviour of
>       this compiler.
>     * But some tools are available to convert the standard GSL headers to
>       MSVC-compatible headers; these tools are based on Perl.
>     * So it is possible to compile, install and use GSL on Windows; but it
>       is definitely not for beginners.
>     * It would be nice to provide a detailed compilation manual or a
>       precompiled binary version somewhere on ipol.im <http://ipol.im>.
>
>     So, if there is a consensus on it, I think we could allow GSL as an
>     external library and mention it in the software guidelines. And if
>     someone has the time to prepare a manual or a precompiled version,
>     it's even better, but should not stop us from accepting GSL.
>
>     --
>     Nicolas LIMARE - CMLA - ENS Cachan http://limare.perso.math.cnrs.fr/
>     IPOL journal http://www.ipol.im/
>     -> image processing, reproducible research, open science
>
>     --
>     IPOL - Image Processing On Line   - http://ipol.im/
>
>     contact edit at ipol.im <mailto:edit at ipol.im>          -
>     http://www.ipol.im/meta/contact/
>     news+feeds  twitter @IPOL_journal - http://www.ipol.im/meta/feeds/
>     announces announce at list.ipol.im <mailto:announce at list.ipol.im> -
>     http://tools.ipol.im/mm/announce/
>     discussions discuss at list.ipol.im <mailto:discuss at list.ipol.im>  -
>     http://tools.ipol.im/mm/discuss/
>
>
>
> --
> IPOL - Image Processing On Line   - http://ipol.im/
>
> contact     edit at ipol.im          - http://www.ipol.im/meta/contact/
> news+feeds  twitter @IPOL_journal - http://www.ipol.im/meta/feeds/
> announces   announce at list.ipol.im - http://tools.ipol.im/mm/announce/
> discussions discuss at list.ipol.im  - http://tools.ipol.im/mm/discuss/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tools.ipol.im/mailman/archive/discuss/attachments/20130826/1bb0a339/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gsl_vs.pdf
Type: application/pdf
Size: 80699 bytes
Desc: not available
URL: <https://tools.ipol.im/mailman/archive/discuss/attachments/20130826/1bb0a339/attachment-0001.pdf>


More information about the discuss mailing list