pygsl-discuss — discussion about the future of pygsl, support

Re: [pygsl-discuss] problem in pygsl library -> building on windows

[Schnizer, P. <pie...@he...>]

Building the package on windows is not trivial.

Building on windows could be documented in a better manner. I’ll give you a collection of resources:

1.       Notes for distributors https://sourceforge.net/p/pygsl/GitRepro/ci/master/tree/doc/ref/readme_distributors.rst

2.       Helper scripts for building on windows https://sourceforge.net/p/pygsl/GitRepro/ci/master/tree/doc/win/

I’ll try to give you some explanations:

1.       Get Msys2. Build GSL using https://sourceforge.net/p/pygsl/GitRepro/ci/master/tree/doc/win/pygsl_msys2_prepare.sh

2.       Change to the pygsl source dir. Then

a.       Copy file gsl_dist / gsl_site_example.py to gsl_dist / gsl_site.py

b.      Edit it to the paths on your machine to GSL and swig

3.       Execute compilation using https://sourceforge.net/p/pygsl/GitRepro/ci/master/tree/doc/win/pygsl_windows_compile.bat
That’s about it
            Pierre


Von: mehrshad ghorbanzadeh [mailto:meh...@gm...]
Gesendet: Mittwoch, 30. Oktober 2019 12:53
An: Schnizer, Pierre <pie...@he...>
Betreff: Re: problem in pygsl library

yes. I have windows 7 (32 bit) on my PC.

On Wed, Oct 30, 2019 at 12:16 PM Schnizer, Pierre <pie...@he...<mailto:pie...@he...>> wrote:
I guess you are using windows.

Is this correct?

Piere

Von: mehrshad ghorbanzadeh [mailto:meh...@gm...<mailto:meh...@gm...>]
Gesendet: Dienstag, 29. Oktober 2019 20:29
An: Ach...@us...<mailto:Ach...@us...>; sch...@us...<mailto:sch...@us...>
Betreff: problem in pygsl library

please fix this error :
"" Traceback (most recent call last):
  File "gsl_dist\gsl_Location.py", line 141, in __init__
    import gsl_site
ModuleNotFoundError: No module named 'gsl_site'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Python36\zpygsl-2.3.0\setup.py", line 99, in <module>
    from gsl_Extension import gsl_Extension
  File "gsl_dist\gsl_Extension.py", line 22, in <module>
    from gsl_Location import gsl_Location
  File "gsl_dist\gsl_Location.py", line 160, in <module>
    gsl_Location = _gsl_Location_file()
  File "gsl_dist\gsl_Location.py", line 148, in __init__
    raise DistutilsExecError(msg % des)
distutils.errors.DistutilsExecError: I do not know how to run gsl-config
on this system. Therefore you must provide me with the information
where to find the GSL library. I could not import `gsl_site'.
Reason: No module named 'gsl_site'. Copy gsl_site_example.py to gsl_site.py.
Edit the variables in that file to reflect your installation.  ""
thanks.

________________________________

Helmholtz-Zentrum Berlin für Materialien und Energie GmbH

Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V.

Aufsichtsrat: Vorsitzender Dr. Volkmar Dietz, stv. Vorsitzende Dr. Jutta Koch-Unterseher
Geschäftsführung: Prof. Dr. Bernd Rech (Sprecher), Prof. Dr. Jan Lüning, Thomas Frederking

Sitz Berlin, AG Charlottenburg, 89 HRB 5583

Postadresse:
Hahn-Meitner-Platz 1
D-14109 Berlin

________________________________

Helmholtz-Zentrum Berlin für Materialien und Energie GmbH

Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V.

Aufsichtsrat: Vorsitzender Dr. Volkmar Dietz, stv. Vorsitzende Dr. Jutta Koch-Unterseher
Geschäftsführung: Prof. Dr. Bernd Rech (Sprecher), Prof. Dr. Jan Lüning, Thomas Frederking

Sitz Berlin, AG Charlottenburg, 89 HRB 5583

Postadresse:
Hahn-Meitner-Platz 1
D-14109 Berlin

Re: [pygsl-discuss] Consult about pyGSL

[Schnizer, P. <pie...@he...>]

multifit and multifit_robust is available in the package

multifit_large are at least available in git.

Please do not hesitate to mail again if that info is not sufficient

Sincerely yours
   Pierre


________________________________
From: luis navarro [lui...@gm...]
Sent: Wednesday, November 28, 2018 8:56 PM
To: Schnizer, Pierre
Cc: sch...@us...; pyg...@li...
Subject: Re: Consult about pyGSL

Hi Pierre,

For sure, this is the simpler interface for N-dimensional least squares fits that was developed by Patrick Alken. I have seen there are some similar functions already wrapped inside pygsl but I was wondering if that one is somewhere within pygsl as well.

Thanks for your help.
[https://mailfoogae.appspot.com/t?sender=abHVpcy5uYXZhcnJvLmRvbWluZ3VlekBnbWFpbC5jb20%3D&type=zerocontent&guid=915c93a5-82eb-4045-94e1-fe1d279e0a8b]ᐧ
[https://mailfoogae.appspot.com/t?sender=abHVpcy5uYXZhcnJvLmRvbWluZ3VlekBnbWFpbC5jb20%3D&type=zerocontent&guid=b7fd6d0a-ce19-4bd8-b840-09cb43634d9f]ᐧ

On Wed, Nov 28, 2018 at 11:42 AM Schnizer, Pierre <pie...@he...<mailto:pie...@he...>> wrote:
Dear Luis,

                please could you clarify to which ndlinear you are refereeing? Fit, mulftifit etc …? Quite some functionality could be available.

Pierre

Von: luis navarro [mailto:lui...@gm...<mailto:lui...@gm...>]
Gesendet: Mittwoch, 28. November 2018 12:07
An: sch...@us...<mailto:sch...@us...>
Betreff: Consult about pyGSL

Dear Schnizer,

Hope you are doing well. I was writing for a quick consult regarding pygsl. I would like to use ndlinear toolbox inside gsl  with python so I was wondering if it is contained within pygsl. I was looking inside the code and I couldn't find it but just any advice from your side would be greatly appreciated.

Thanks for your time.
--
Luis Navarro Domínguez
[https://mailfoogae.appspot.com/t?sender=abHVpcy5uYXZhcnJvLmRvbWluZ3VlekBnbWFpbC5jb20%3D&type=zerocontent&guid=d9ff6e4c-1dba-47a3-bb39-db646bad1732]ᐧ

________________________________

Helmholtz-Zentrum Berlin für Materialien und Energie GmbH

Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V.

Aufsichtsrat: Vorsitzender Dr. Karl Eugen Huthmacher, stv. Vorsitzende Dr. Jutta Koch-Unterseher
Geschäftsführung: Prof. Dr. Bernd Rech (kommissarisch), Thomas Frederking

Sitz Berlin, AG Charlottenburg, 89 HRB 5583

Postadresse:
Hahn-Meitner-Platz 1
D-14109 Berlin


--
Luis Navarro Domínguez

________________________________

Helmholtz-Zentrum Berlin für Materialien und Energie GmbH

Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V.

Aufsichtsrat: Vorsitzender Dr. Karl Eugen Huthmacher, stv. Vorsitzende Dr. Jutta Koch-Unterseher
Geschäftsführung: Prof. Dr. Bernd Rech (kommissarisch), Thomas Frederking

Sitz Berlin, AG Charlottenburg, 89 HRB 5583

Postadresse:
Hahn-Meitner-Platz 1
D-14109 Berlin

Re: [pygsl-discuss] Consult about pyGSL

[Schnizer, P. <pie...@he...>]

Dear Luis,

                please could you clarify to which ndlinear you are refereeing? Fit, mulftifit etc …? Quite some functionality could be available.

Pierre

Von: luis navarro [mailto:lui...@gm...]
Gesendet: Mittwoch, 28. November 2018 12:07
An: sch...@us...
Betreff: Consult about pyGSL

Dear Schnizer,

Hope you are doing well. I was writing for a quick consult regarding pygsl. I would like to use ndlinear toolbox inside gsl  with python so I was wondering if it is contained within pygsl. I was looking inside the code and I couldn't find it but just any advice from your side would be greatly appreciated.

Thanks for your time.
--
Luis Navarro Domínguez
[https://mailfoogae.appspot.com/t?sender=abHVpcy5uYXZhcnJvLmRvbWluZ3VlekBnbWFpbC5jb20%3D&type=zerocontent&guid=d9ff6e4c-1dba-47a3-bb39-db646bad1732]ᐧ

________________________________

Helmholtz-Zentrum Berlin für Materialien und Energie GmbH

Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V.

Aufsichtsrat: Vorsitzender Dr. Karl Eugen Huthmacher, stv. Vorsitzende Dr. Jutta Koch-Unterseher
Geschäftsführung: Prof. Dr. Bernd Rech (kommissarisch), Thomas Frederking

Sitz Berlin, AG Charlottenburg, 89 HRB 5583

Postadresse:
Hahn-Meitner-Platz 1
D-14109 Berlin

Re: [pygsl-discuss] Oldest GSL version in the wilde

[José A. M. <jao...@gm...>]

On Saturday, 1 September 2018 13.38.43 WEST Schnizer, Pierre wrote:
> Dear all,
> 
>                 I am currently revising the configuration code so that it
> will be compatible with pip. Now that I will test the code, I’d like to
> know which is the oldest version  of GSL still in use.
> 
> Please could you indicate:
> 
> ·         GSL 1.16?
> ·         GSL older than 1.16?
> ·         GSL 2.0: has a bug in multifit, pygsl does currently not support
> this released
> ·         GSL 2.1?
> 
> Any response is highly appretiated!
> 
> Sincerely yours
> 
>                 Pierre

With my Fedora hat I can say that the oldest version from the distribution 
family is that from "Red Hat Enterprise Linux 7" that has gsl 1.15:
https://centos.pkgs.org/7/centos-x86_64/gsl-1.15-13.el7.x86_64.rpm.html

Regards,
-- 
José Matos

[pygsl-discuss] Oldest GSL version in the wilde

[Schnizer, P. <pie...@he...>]

Dear all,

                I am currently revising the configuration code so that it will be compatible with pip. Now that I will test the code, I’d like to know which is the oldest version  of GSL still in use.

Please could you indicate:

·         GSL 1.16?

·         GSL older than 1.16?

·         GSL 2.0: has a bug in multifit, pygsl does currently not support this released

·         GSL 2.1?

Any response is highly appretiated!

Sincerely yours
                Pierre

________________________________

Helmholtz-Zentrum Berlin für Materialien und Energie GmbH

Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V.

Aufsichtsrat: Vorsitzender Dr. Karl Eugen Huthmacher, stv. Vorsitzende Dr. Jutta Koch-Unterseher
Geschäftsführung: Prof. Dr. Bernd Rech (kommissarisch), Thomas Frederking

Sitz Berlin, AG Charlottenburg, 89 HRB 5583

Postadresse:
Hahn-Meitner-Platz 1
D-14109 Berlin

https://www.helmholtz-berlin.de

Abonnieren Sie unseren monatlichen Newsletter!<https://www.helmholtz-berlin.de/aktuell/pm/newsletter/>
Subscribe to our monthly newsletter!<https://www.helmholtz-berlin.de/aktuell/pm/newsletter/index_en.html>

[pygsl-discuss] GIT fails to build : missing pygsl.errno

[Helmut J. <jar...@sk...>]

Hi,

setup.py fails since it requires errno e.g.
pygsl/errors.py : from pygsl import errno as errno_m

but there is no errno.py anywhere.

Is there a patch?

Thanks,
Helmut

Re: [pygsl-discuss] PyPI and conda-forge package for PyGSL 2.2

[Marcel S. <mar...@in...>]

Hi again,

apparently PyGSL seems still to be actively developed (version 2.3 has 
been released by now) -- I'm still interested in responses to my 
questions below.

Thanks, best

   Marcel


On 29/08/17 19:09, Marcel Stimberg wrote:
> Hi everyone,
>
> I've seen on SF that PyGSL 2.2 has been released, but it does not seem 
> to have been uploaded to PyPI[1]. Is this something you were planning 
> to do? I also had a cursory look at bundling PyGSL into a conda 
> package for upload to conda-forge[2] -- is this something that you 
> would be interested in? IMHO, it could be very useful given that the 
> anaconda distribution is now widely used in academic circles, and 
> supporting installation of packages that have non-Python dependencies 
> (e.g. the GSL in this case) was one of the major reasons behind 
> creating it.
>
> If you are interested, I'm happy to give it a go (in that case it 
> would be great if one or several of the maintainers could send me 
> their github handle so that I can add them as maintainers to the conda 
> package).
>
> Best,
>
>   Marcel
>
> [1] https://pypi.python.org/pypi/pygsl/2.1.1
>
> [2] https://conda-forge.org/#about
>

[pygsl-discuss] PyPI and conda-forge package for PyGSL 2.2

[Marcel S. <mar...@in...>]

Hi everyone,

I've seen on SF that PyGSL 2.2 has been released, but it does not seem 
to have been uploaded to PyPI[1]. Is this something you were planning to 
do? I also had a cursory look at bundling PyGSL into a conda package for 
upload to conda-forge[2] -- is this something that you would be 
interested in? IMHO, it could be very useful given that the anaconda 
distribution is now widely used in academic circles, and supporting 
installation of packages that have non-Python dependencies (e.g. the GSL 
in this case) was one of the major reasons behind creating it.

If you are interested, I'm happy to give it a go (in that case it would 
be great if one or several of the maintainers could send me their github 
handle so that I can add them as maintainers to the conda package).

Best,

   Marcel

[1] https://pypi.python.org/pypi/pygsl/2.1.1

[2] https://conda-forge.org/#about

Re: [pygsl-discuss] PyGSL 2.2.0 released

[Jonathan S. <jjs...@gm...>]

Thanks. I have installed successfully for Python 2.7 on Mac OS with
Macports dependencies.

Jonathan


On Thu, Apr 28, 2016 at 11:40 AM, Pierre Schnizer <p.s...@gs...> wrote:

> Dear Jonathan,
>
>         the errors in the config are ok and should not stop the process.
>
> GSL 2.1 defines the functions gsl_sf_legendre_Plm_array
> gsl_sf_legendre_sphPlm_array
> in the headers but are not found in the library object. These tests below
> should fail in 2.1
> Then you should not find the macros
> _PYGSL_GSL_HAS_LINK_GSL_SF_LEGENDRE_PLM_ARRAY
> _PYGSL_GSL_HAS_LINK_GSL_SF_LEGENDRE_SPHPLM_ARRAY
> in your Include/pygsl/pygsl_features_config.h
>
> These are however still available in e.g. GSL 1.16. and should be part of
> the module
> pygsl.testing.sf if pygsl is compiled against GSL-1.16
>
> Sincerely yours
>        Pierre
>
>
>
> Am 28.04.2016 um 19:21 schrieb Jonathan Stickel:
>
> > On 4/18/16 12:54 , Pierre Schnizer wrote:
> >> Dear All,
> >>
> >>      a new release was uploaded to SF.
> >> This release was only tested with gsl-2.1.
> >>
> >> Sincerely yours
> >>      Pierre
> >>
> >>
> >
> >
> > Thanks for the release. I am trying to compile on Mac, using
> > dependencies from Macports. Doing
> >
> > $ python setup.py config
> >
> > results in these errors:
> >
> >
> > compiling '_configtest.c':
> > #include <gsl/gsl_sf.h>
> >
> > int main () {
> >   gsl_sf_legendre_Plm_array(0, 0, 0.0, NULL);
> > }
> >
> > C compiler: /usr/bin/clang -fno-strict-aliasing -fno-common -dynamic
> > -pipe -Os -fwrapv -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes
> >
> > compile options: '-I/opt/local/include -c'
> > clang: _configtest.c
> > /usr/bin/clang _configtest.o -L/opt/local/lib -lgsl -lgslcblas -o
> > _configtest
> > Undefined symbols for architecture x86_64:
> >   "_gsl_sf_legendre_Plm_array", referenced from:
> >       _main in _configtest.o
> > ld: symbol(s) not found for architecture x86_64
> > clang: error: linker command failed with exit code 1 (use -v to see
> > invocation)
> > Undefined symbols for architecture x86_64:
> >   "_gsl_sf_legendre_Plm_array", referenced from:
> >       _main in _configtest.o
> > ld: symbol(s) not found for architecture x86_64
> > clang: error: linker command failed with exit code 1 (use -v to see
> > invocation)
> > failure.
> > removing: _configtest.c _configtest.o
> > compiling '_configtest.c':
> > #include <gsl/gsl_sf.h>
> >
> > int main () {
> >   gsl_sf_legendre_sphPlm_array(0, 0, 0.0, NULL);
> > }
> >
> > C compiler: /usr/bin/clang -fno-strict-aliasing -fno-common -dynamic
> > -pipe -Os -fwrapv -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes
> >
> > compile options: '-I/opt/local/include -c'
> > clang: _configtest.c
> > /usr/bin/clang _configtest.o -L/opt/local/lib -lgsl -lgslcblas -o
> > _configtest
> > Undefined symbols for architecture x86_64:
> >   "_gsl_sf_legendre_sphPlm_array", referenced from:
> >       _main in _configtest.o
> > ld: symbol(s) not found for architecture x86_64
> > clang: error: linker command failed with exit code 1 (use -v to see
> > invocation)
> > Undefined symbols for architecture x86_64:
> >   "_gsl_sf_legendre_sphPlm_array", referenced from:
> >       _main in _configtest.o
> > ld: symbol(s) not found for architecture x86_64
> > clang: error: linker command failed with exit code 1 (use -v to see
> > invocation)
> > failure.
> > removing: _configtest.c _configtest.o
> >
> >
> > Do you have a suggestion?
> >
> > Thanks,
> > Jonathan
> >
> >
> ------------------------------------------------------------------------------
> > Find and fix application performance issues faster with Applications
> Manager
> > Applications Manager provides deep performance insights into multiple
> tiers of
> > your business applications. It resolves application problems quickly and
> > reduces your MTTR. Get your free trial!
> > https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
> > _______________________________________________
> > pygsl-discuss mailing list
> > pyg...@li...
> > https://lists.sourceforge.net/lists/listinfo/pygsl-discuss
>
>

Re: [pygsl-discuss] PyGSL 2.2.0 released

[Pierre S. <p.s...@gs...>]

Dear Jonathan,

	the errors in the config are ok and should not stop the process.

GSL 2.1 defines the functions gsl_sf_legendre_Plm_array gsl_sf_legendre_sphPlm_array
in the headers but are not found in the library object. These tests below should fail in 2.1
Then you should not find the macros 
_PYGSL_GSL_HAS_LINK_GSL_SF_LEGENDRE_PLM_ARRAY
_PYGSL_GSL_HAS_LINK_GSL_SF_LEGENDRE_SPHPLM_ARRAY
in your Include/pygsl/pygsl_features_config.h 

These are however still available in e.g. GSL 1.16. and should be part of the module
pygsl.testing.sf if pygsl is compiled against GSL-1.16

Sincerely yours
       Pierre

 

Am 28.04.2016 um 19:21 schrieb Jonathan Stickel:

> On 4/18/16 12:54 , Pierre Schnizer wrote:
>> Dear All,
>> 
>> 	a new release was uploaded to SF.
>> This release was only tested with gsl-2.1.
>> 
>> Sincerely yours
>> 	Pierre
>> 
>> 
> 
> 
> Thanks for the release. I am trying to compile on Mac, using 
> dependencies from Macports. Doing
> 
> $ python setup.py config
> 
> results in these errors:
> 
> 
> compiling '_configtest.c':
> #include <gsl/gsl_sf.h>
> 
> int main () {
>   gsl_sf_legendre_Plm_array(0, 0, 0.0, NULL);
> }
> 
> C compiler: /usr/bin/clang -fno-strict-aliasing -fno-common -dynamic 
> -pipe -Os -fwrapv -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes
> 
> compile options: '-I/opt/local/include -c'
> clang: _configtest.c
> /usr/bin/clang _configtest.o -L/opt/local/lib -lgsl -lgslcblas -o 
> _configtest
> Undefined symbols for architecture x86_64:
>   "_gsl_sf_legendre_Plm_array", referenced from:
>       _main in _configtest.o
> ld: symbol(s) not found for architecture x86_64
> clang: error: linker command failed with exit code 1 (use -v to see 
> invocation)
> Undefined symbols for architecture x86_64:
>   "_gsl_sf_legendre_Plm_array", referenced from:
>       _main in _configtest.o
> ld: symbol(s) not found for architecture x86_64
> clang: error: linker command failed with exit code 1 (use -v to see 
> invocation)
> failure.
> removing: _configtest.c _configtest.o
> compiling '_configtest.c':
> #include <gsl/gsl_sf.h>
> 
> int main () {
>   gsl_sf_legendre_sphPlm_array(0, 0, 0.0, NULL);
> }
> 
> C compiler: /usr/bin/clang -fno-strict-aliasing -fno-common -dynamic 
> -pipe -Os -fwrapv -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes
> 
> compile options: '-I/opt/local/include -c'
> clang: _configtest.c
> /usr/bin/clang _configtest.o -L/opt/local/lib -lgsl -lgslcblas -o 
> _configtest
> Undefined symbols for architecture x86_64:
>   "_gsl_sf_legendre_sphPlm_array", referenced from:
>       _main in _configtest.o
> ld: symbol(s) not found for architecture x86_64
> clang: error: linker command failed with exit code 1 (use -v to see 
> invocation)
> Undefined symbols for architecture x86_64:
>   "_gsl_sf_legendre_sphPlm_array", referenced from:
>       _main in _configtest.o
> ld: symbol(s) not found for architecture x86_64
> clang: error: linker command failed with exit code 1 (use -v to see 
> invocation)
> failure.
> removing: _configtest.c _configtest.o
> 
> 
> Do you have a suggestion?
> 
> Thanks,
> Jonathan
> 
> ------------------------------------------------------------------------------
> Find and fix application performance issues faster with Applications Manager
> Applications Manager provides deep performance insights into multiple tiers of
> your business applications. It resolves application problems quickly and
> reduces your MTTR. Get your free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
> _______________________________________________
> pygsl-discuss mailing list
> pyg...@li...
> https://lists.sourceforge.net/lists/listinfo/pygsl-discuss

Re: [pygsl-discuss] PyGSL 2.2.0 released

[Jonathan S. <jjs...@gm...>]

On 4/18/16 12:54 , Pierre Schnizer wrote:
> Dear All,
>
> 	a new release was uploaded to SF.
> This release was only tested with gsl-2.1.
>
> Sincerely yours
> 	Pierre
>
>


Thanks for the release. I am trying to compile on Mac, using 
dependencies from Macports. Doing

$ python setup.py config

results in these errors:


compiling '_configtest.c':
#include <gsl/gsl_sf.h>

int main () {
   gsl_sf_legendre_Plm_array(0, 0, 0.0, NULL);
}

C compiler: /usr/bin/clang -fno-strict-aliasing -fno-common -dynamic 
-pipe -Os -fwrapv -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes

compile options: '-I/opt/local/include -c'
clang: _configtest.c
/usr/bin/clang _configtest.o -L/opt/local/lib -lgsl -lgslcblas -o 
_configtest
Undefined symbols for architecture x86_64:
   "_gsl_sf_legendre_Plm_array", referenced from:
       _main in _configtest.o
ld: symbol(s) not found for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see 
invocation)
Undefined symbols for architecture x86_64:
   "_gsl_sf_legendre_Plm_array", referenced from:
       _main in _configtest.o
ld: symbol(s) not found for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see 
invocation)
failure.
removing: _configtest.c _configtest.o
compiling '_configtest.c':
#include <gsl/gsl_sf.h>

int main () {
   gsl_sf_legendre_sphPlm_array(0, 0, 0.0, NULL);
}

C compiler: /usr/bin/clang -fno-strict-aliasing -fno-common -dynamic 
-pipe -Os -fwrapv -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes

compile options: '-I/opt/local/include -c'
clang: _configtest.c
/usr/bin/clang _configtest.o -L/opt/local/lib -lgsl -lgslcblas -o 
_configtest
Undefined symbols for architecture x86_64:
   "_gsl_sf_legendre_sphPlm_array", referenced from:
       _main in _configtest.o
ld: symbol(s) not found for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see 
invocation)
Undefined symbols for architecture x86_64:
   "_gsl_sf_legendre_sphPlm_array", referenced from:
       _main in _configtest.o
ld: symbol(s) not found for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see 
invocation)
failure.
removing: _configtest.c _configtest.o


Do you have a suggestion?

Thanks,
Jonathan

Re: [pygsl-discuss] PyGSL 2.2.0 released

[José M. <jao...@gm...>]

On Monday, April 18, 2016 8:54:50 PM WEST Pierre Schnizer wrote:
> Dear All,
> 
>         a new release was uploaded to SF.
> This release was only tested with gsl-2.1. 
> 
> Sincerely yours
>         Pierre

Thank you Pierre.
I have built the package for Fedora 24 and development without issues.

Regards,
-- 
José Matos

[pygsl-discuss] PyGSL 2.2.0 released

[Pierre S. <p.s...@gs...>]

Dear All,

	a new release was uploaded to SF.
This release was only tested with gsl-2.1. 

Sincerely yours
	Pierre

Re: [pygsl-discuss] upcoming new release

[Markus M. <mar...@zo...>]

Thanks for the notification. Looking forward to the new release.



Markus

---- On Wed, 13 Apr 2016 17:25:50 +1200 Pierre Schnizer <p.s...@gs...> wrote ---- 

Dear all, 
 Thanks to all for your interest. I plan to release a new  version during this weekend.


--  I fixed errortestmodule.c  in CVS
--  multifit_robust is in CVS not yet in the last release (2.1.1)


Sincerely yours
 Pierre


Am 13.04.2016 um 04:18 schrieb Markus Müller:

 
Dear listers,

I need to use the GSL multifit robust regression routine. I just installed GSL 1.16 and pygsl 2.1.1. I can import pygsl.multifit, into python but pygsl.multifit.robust is not available (e.g. help(pygsl.multifit.robust) fails with an AttributeError). My first conclusion was that robust regression is not yet available via pygsl, but then I found pygsl/swig_src/multifit_robust_warp.c in pygsl bitbucket repo. I therefore assume the needed functionality is implemented.


What could I have done wrong? Any help appreciated.




Cheers,


Markus






------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z_______________________________________________
pygsl-discuss mailing list
pyg...@li...
https://lists.sourceforge.net/lists/listinfo/pygsl-discuss




------------------------------------------------------------------------------ 
Find and fix application performance issues faster with Applications Manager 
Applications Manager provides deep performance insights into multiple tiers of 
your business applications. It resolves application problems quickly and 
reduces your MTTR. Get your free trial! 
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z_______________________________________________ 
pygsl-discuss mailing list 
pyg...@li... 
https://lists.sourceforge.net/lists/listinfo/pygsl-discuss 

[pygsl-discuss] upcoming new release

[Pierre S. <p.s...@gs...>]

Dear all,
	
	Thanks to all for your interest. I plan to release a new  version during this weekend.

--  I fixed errortestmodule.c  in CVS
--  multifit_robust is in CVS not yet in the last release (2.1.1)

Sincerely yours
	Pierre

Am 13.04.2016 um 04:18 schrieb Markus Müller:

> 
> Dear listers,
> 
> I need to use the GSL multifit robust regression routine. I just installed GSL 1.16 and pygsl 2.1.1. I can import pygsl.multifit, into python but pygsl.multifit.robust is not available (e.g. help(pygsl.multifit.robust) fails with an AttributeError). My first conclusion was that robust regression is not yet available via pygsl, but then I found pygsl/swig_src/multifit_robust_warp.c in pygsl bitbucket repo. I therefore assume the needed functionality is implemented.
> 
> What could I have done wrong? Any help appreciated.
> 
> 
> Cheers,
> 
> Markus
> 
> 
> ------------------------------------------------------------------------------
> Find and fix application performance issues faster with Applications Manager
> Applications Manager provides deep performance insights into multiple tiers of
> your business applications. It resolves application problems quickly and
> reduces your MTTR. Get your free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z_______________________________________________
> pygsl-discuss mailing list
> pyg...@li...
> https://lists.sourceforge.net/lists/listinfo/pygsl-discuss

[pygsl-discuss] Support for multi-variate robust regression routine

[Markus M. <mar...@zo...>]

Dear listers,

I need to use the GSL multifit robust regression routine. I just installed GSL 1.16 and pygsl 2.1.1. I can import pygsl.multifit, into python but pygsl.multifit.robust is not available (e.g. help(pygsl.multifit.robust) fails with an AttributeError). My first conclusion was that robust regression is not yet available via pygsl, but then I found pygsl/swig_src/multifit_robust_warp.c in pygsl bitbucket repo. I therefore assume the needed functionality is implemented.


What could I have done wrong? Any help appreciated.




Cheers,


Markus

[pygsl-discuss] errortestmodule.c returns with value from void function (Python 2.x)

[Cong Ma <fri...@gm...>]

When compiling with Python 2.x, the function void initerrortest in
src/errortestmodule.c returns NULL under label "fail".  This prevents
the code from compiling because returning anything from void is
impossible.

Could this be fixed?

Best wishes,
Cong.

[pygsl-discuss] release for gsl-2.1

[Jonathan S. <jjs...@gm...>]

Hey,

 From browsing the email archives, it looks like support for gsl-2.1 is 
available in CVS. It would be great if a release was snapshot was made. 
This will help with packaging for various distribution systems. My 
interest is for Macports.

Thanks!
Jonathan

Re: [pygsl-discuss] PyGSL fails to build from source

[Cong Ma <fri...@gm...>]

On Mon, Apr 4, 2016 at 11:43 AM, Pierre Schnizer <p.s...@gs...> wrote:
> Dear Cong,
>
>         please can you send me the file:
>
> Include/pygsl/pygsl_features_config.h

Here it is in the attached file.

The relevant code section in swig_src/callback_wrap.c is as follows

"""
#include <pygsl/pygsl_features.h>
#ifdef _PYGSL_GSL_HAS_FMINIMIZER_NMSIMPLEX
      extern const
           gsl_multimin_fminimizer_type *gsl_multimin_fminimizer_nmsimplex;
#else
      gsl_multimin_fminimizer_type *gsl_multimin_fminimizer_nmsimplex = NULL;
#endif
"""

On inspecting the file you requested, could it be that the macro to
check should be _PYGSL_GSL_HAS_MULTIMIN_FMINIMIZER_NMSIMPLEX, rather
than _PYGSL_GSL_HAS_FMINIMIZER_NMSIMPLEX as in the quoted code section
above?

Best wishes,
Cong.

Re: [pygsl-discuss] PyGSL fails to build from source

[Pierre S. <p.s...@gs...>]

Dear Cong,

	please can you send me the file:

Include/pygsl/pygsl_features_config.h 

Thanks,
	Pierre


Am 04.04.2016 um 11:36 schrieb Cong Ma:

> Dear all,
> 
> I have trouble building the recent (2016-04) source from CVS.  The GSL
> version is 1.16.  I tried with both Linux and Mac OS X (with GSL is
> installed from the homebrew repository for the latter), and the error
> is the same.
> 
> The "python setup.py config" step seems to go fine.  The following
> error is encountered during the build stage:
> 
> """
> swig_src/callback_wrap.c:4149:34: error: redefinition of
> 'gsl_multimin_fminimizer_nmsimplex' with a different type:
> 'gsl_multimin_fminimizer_type *' vs 'const
> gsl_multimin_fminimizer_type *'
>          gsl_multimin_fminimizer_type
> *gsl_multimin_fminimizer_nmsimplex = NULL;
>                                        ^
> /usr/local/Cellar/gsl/1.16/include/gsl/gsl_multimin.h:220:45: note:
> previous declaration is here
> GSL_VAR const gsl_multimin_fminimizer_type *gsl_multimin_fminimizer_nmsimplex;
> """
> 
> It seems that the configured PyGSL source falsely believes that
> gsl_multimin_fminimizer_nmsimplex is not available (?)
> 
> Best wishes,
> Cong.
> 
> ------------------------------------------------------------------------------
> _______________________________________________
> pygsl-discuss mailing list
> pyg...@li...
> https://lists.sourceforge.net/lists/listinfo/pygsl-discuss

[pygsl-discuss] PyGSL fails to build from source

[Cong Ma <fri...@gm...>]

Dear all,

I have trouble building the recent (2016-04) source from CVS.  The GSL
version is 1.16.  I tried with both Linux and Mac OS X (with GSL is
installed from the homebrew repository for the latter), and the error
is the same.

The "python setup.py config" step seems to go fine.  The following
error is encountered during the build stage:

"""
swig_src/callback_wrap.c:4149:34: error: redefinition of
'gsl_multimin_fminimizer_nmsimplex' with a different type:
'gsl_multimin_fminimizer_type *' vs 'const
gsl_multimin_fminimizer_type *'
          gsl_multimin_fminimizer_type
*gsl_multimin_fminimizer_nmsimplex = NULL;
                                        ^
/usr/local/Cellar/gsl/1.16/include/gsl/gsl_multimin.h:220:45: note:
previous declaration is here
GSL_VAR const gsl_multimin_fminimizer_type *gsl_multimin_fminimizer_nmsimplex;
"""

It seems that the configured PyGSL source falsely believes that
gsl_multimin_fminimizer_nmsimplex is not available (?)

Best wishes,
Cong.

Re: [pygsl-discuss] New release

[José M. <jao...@gm...>]

On Sunday, March 13, 2016 11:29:44 AM Pierre SCHNIZER wrote:
> Dear Jose,
> 
>        cvs should now compile for gsl1.16 and gsl2.1 
> Could you check?
> 
> Thx
>     Pierre

It does. :-D

I checked with Fedora 23 (gsl-1.16) and Fedora 24 (gsl-2.1).

Thank you. :-)
-- 
José Matos

Re: [pygsl-discuss] New release

[Pierre S. <p.s...@gs...>]

Dear Jose,

       cvs should now compile for gsl1.16 and gsl2.1 
Could you check?

Thx
    Pierre

On 09.03.2016 12:55, José Matos wrote:
> On Wednesday, March 09, 2016 09:31:24 AM Schnizer, Pierre Dr. wrote:
>> Dear Jose,
>> 
>>    the code in CVS should compile now for GSL-2.1.
>> 
>> Could you give it a try?
>> Pierre
> I am struggling with it. :-)
> 
> I do not have yet a machine with (development version of what it will be) 
> Fedora 24. And I am using a chroot to build it.
> 
> Issues:
> 1) When building on Fedora 23 with gsl-1.16 and gcc 5.3 (with python 2.7):
> 
> swig_src/callback_wrap.c: In function 
> ‘Swig_var_gsl_multifit_fdfsolver_lmniel_get’:
> swig_src/callback_wrap.c:10841:67: error: ‘gsl_multifit_fdfsolver_lmniel’ 
> undeclared (first use in this function)
>    pyobj = SWIG_NewPointerObj(SWIG_as_voidptr(gsl_multifit_fdfsolver_lmniel), 
> SWIGTYPE_p_gsl_multifit_fdfsolver_type,  0 );
> 
> 2) When building on the development version with gsl-2.1 and (to be) gcc 6 
> (with python 2.7):
> testing/src/solvers/multifit_nlin.c: In function 
> 'PyGSL_multifit_fdfsolver_test_gradient':
> testing/src/solvers/multifit_nlin.c:128:36: error: 'gsl_multifit_fdfsolver 
> {aka struct <anonymous>}' has no member named 'J'
>       flag = gsl_multifit_gradient(s->J, s->f, g);
> 
> 
> 3) There seems to be a typo in gslwrap/math.i (patch attached at pygsl-fix-
> pragma.patch)
> 
> Regards,
> 
> 
> ------------------------------------------------------------------------------
> Transform Data into Opportunity.
> Accelerate data analysis in your applications with
> Intel Data Analytics Acceleration Library.
> Click to learn more.
> http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
> 
> 
> _______________________________________________
> pygsl-discuss mailing list
> pyg...@li...
> https://lists.sourceforge.net/lists/listinfo/pygsl-discuss

-- 
+---------------------------------------------------------------------+
  Pierre Schnizer  <p.s...@gs...>
  Telephon : +49 6159 71 1557
  Fax      : +49 6159 71 3099

GSI Helmholtzzentrum für Schwerionenforschung GmbH
Planckstraße 1
64291 Darmstadt
www.gsi.de

Gesellschaft mit beschränkter Haftung
Sitz der Gesellschaft: Darmstadt
Handelsregister: Amtsgericht Darmstadt, HRB 1528

Geschäftsführung:
Ursula Weyrich, Prof. Dr. Karlheinz Langanke, Jörg Blaurock

Vorsitzender des Aufsichtsrates: Staatssekretär Dr. Georg Schütte
Stellvertreter: Ministerialdirigent Dr. Rolf Bernhardt
+---------------------------------------------------------------------+

Re: [pygsl-discuss] New release

[José M. <jao...@gm...>]

On Wednesday, March 09, 2016 09:31:24 AM Schnizer, Pierre Dr. wrote:
> Dear Jose,
> 
>    the code in CVS should compile now for GSL-2.1.
> 
> Could you give it a try?
> Pierre

I am struggling with it. :-)

I do not have yet a machine with (development version of what it will be) 
Fedora 24. And I am using a chroot to build it.

Issues:
1) When building on Fedora 23 with gsl-1.16 and gcc 5.3 (with python 2.7):

swig_src/callback_wrap.c: In function 
‘Swig_var_gsl_multifit_fdfsolver_lmniel_get’:
swig_src/callback_wrap.c:10841:67: error: ‘gsl_multifit_fdfsolver_lmniel’ 
undeclared (first use in this function)
   pyobj = SWIG_NewPointerObj(SWIG_as_voidptr(gsl_multifit_fdfsolver_lmniel), 
SWIGTYPE_p_gsl_multifit_fdfsolver_type,  0 );

2) When building on the development version with gsl-2.1 and (to be) gcc 6 
(with python 2.7):
testing/src/solvers/multifit_nlin.c: In function 
'PyGSL_multifit_fdfsolver_test_gradient':
testing/src/solvers/multifit_nlin.c:128:36: error: 'gsl_multifit_fdfsolver 
{aka struct <anonymous>}' has no member named 'J'
      flag = gsl_multifit_gradient(s->J, s->f, g);


3) There seems to be a typo in gslwrap/math.i (patch attached at pygsl-fix-
pragma.patch)

Regards,
-- 
José Matos

Re: [pygsl-discuss] New release

[Schnizer, P. Dr. <P.S...@gs...>]

Dear Jose,

   the code in CVS should compile now for GSL-2.1.

Could you give it a try?
Pierre



José Matos <jao...@gm...> schrieb:


On Monday, March 07, 2016 09:21:07 PM Pierre Schnizer wrote:
> Dear Jose,
>
>         there is a bit more to it … the pygsl_features.h file I uploaded to
> cvs , I hope I can  add the other files tomorrow …
>
> Thanks for pointing it out
>    Pierre

Thank you for taking care of this issue. :-)

Regards,
--
José Matos

------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://makebettercode.com/inteldaal-eval
_______________________________________________
pygsl-discuss mailing list
pyg...@li...
https://lists.sourceforge.net/lists/listinfo/pygsl-discuss