Hey Ed, Daniel,
I pulled the 1.2 tree from svn, installed the Python 2.4 and the Windows
dependancies you listed, including the readline package, and did some
testing. (I also installed TortoiseSVN, gvim and bash to make myself feel a
little more at home).
Using the readline package Ed pointed to
(http://sourceforge.net/project/showfiles.php?group_id=82407&package_id=84552
), it dies with:
-------
File
"C:\Python24\Lib\site-packages\readline\PyReadline.py",
line 12, in ?
import win32con as c32
ImportError: No module named win32con
-------
While we could attempt to debug this (not sure exactly where win32con can be
found ..), there is another solution.
Uninstall the original readline package, and install pyreadline instead (a
"pure python" implementation).
(http://projects.scipy.org/ipython/ipython/wiki/PyReadline/Intro
)
(http://ipython.scipy.org/dist/pyreadline-1.3.win32.exe).
Hopefully the pyreadline implementation is complete enough to work for
relax. It seems to work for my limited testing.
I ran the test-suite, the only failures were:
Mapping the {local_tm, S2, te} chi2 space through the OpenDX user function
dx.map() . [ Failed ]
AIC model selection between two diffusion tensors
................................... [ Failed ]
Synopsis
................................................................................
[ Failed ]
and some Tracebacks:
--------
ImportError: relaxation curve fitting is unavailible, try compiling the C
modules.
Traceback (most recent call last):
File "C:\Documents and
Settings\perry\Desktop\relax-1.2svn\test_suite\test_suite.py", line 251, in
exec_tests
test_array[i].passed = test_array[i].test(self.run )
File "C:\Documents and
Settings\perry\Desktop\relax-1.2svn\test_suite\model_free.py", line 260, in
opendx_tm_s2_te
self.relax.interpreter._OpenDX.map(self.run,
params=['local_tm', 'S2', 'te'], res_num=2, inc=2, file='devnull')
File "C:\Documents and Settings\perry\Desktop\relax-1.2svn\prompt\dx.py",
line 279, in map
self.__relax__.generic.opendx.map(run=run,
params=params, map_type=map_type, res_num=res_num, inc=inc, lower=lower,
upper=upper, axis_in
cs=axis_incs, file=file, dir=dir, point=point, point_file=point_file,
remap=remap)
File "C:\Documents and Settings\perry\Desktop\relax-1.2svn\dx\opendx.py",
line 59, in map
self.iso3d.map_space(run, params, res_num, index, inc, lower, upper,
axis_incs, file, dir, point, point_file, remap)
File "C:\Documents and
Settings\perry\Desktop\relax-1.2svn\dx\base_map.py", line 347, in map_space
self.create_map()
File "C:\Documents and Settings\perry\Desktop\relax-1.2svn\dx\base_map.py
", line 81, in create_map
self.map_text(map_file)
File "C:\Documents and
Settings\perry\Desktop\relax-1.2svn\dx\isosurface_3D.py", line 104, in
map_text
self.calculate(run=self.run, res_num= self.res_num, print_flag=0)
File "C:\Documents and
Settings\perry\Desktop\relax-1.2svn\specific_fns\model_free.py",
line 546, in calculate
chi2 = self.mf.func(self.param_vector)
File "C:\Documents and Settings\perry\Desktop\relax-
1.2svn\maths_fns\mf.py", line 414, in func_mf
data.chi2 = chi2(data.relax_data, data.ri, data.errors)
File "C:\Documents and
Settings\perry\Desktop\relax-1.2svn\maths_fns\chi2.py", line 50, in chi2
return sum((1.0 / errors * (data - back_calc_vals))**2)
ValueError: math domain error
Traceback (most recent call last):
File "C:\Documents and
Settings\perry\Desktop\relax-1.2svn\test_suite\test_suite.py ", line 251, in
exec_tests
test_array[i].passed = test_array[i].test(self.run)
File "C:\Documents and
Settings\perry\Desktop\relax-1.2svn\test_suite\model_selection.py",
line 78, in diff
self.relax.interpreter._Modsel.model_selection(method='AIC',
modsel_run='aic')
File "C:\Documents and
Settings\perry\Desktop\relax-1.2svn\prompt\model_selection.py",
line 121, in model_selection
self.relax.generic.model_selection.select
(method=method, modsel_run=modsel_run, runs=runs)
File "C:\Documents and
Settings\perry\Desktop\relax-1.2svn\generic_fns\model_selection.py",
line 169, in select
best_crit = float('inf')
ValueError: invalid literal for float(): inf
-------
(These may be errors for the current 1.2 SVN version under linux too ... I
haven't checked).
Anyway, hopefully you can get a bit further now ..
Andrew Perry
On 9/4/06, Edward d'Auvergne <edward@xxxxxxxxxxxxx> wrote:
> I've modified the relax download page
> (http://www.nmr-relax.com/download.html) to have
different sections
> for each operating system and to include links to the relax
> dependencies for those platforms. The links under the MS Windows
> section may help in getting relax to run on that platform. Are these
> the files you installed Daniel? It also includes the Python readline
> alternative for Windows. It may take a few hours for
> http://nmr-relax.com to update itself and display the new pages.
>
> The commit message of the changes is located at
>
https://mail.gna.org/public/relax-commits/2006-09/msg00010.html
> (Message-id: <E1GKA5S-0000kV-7I@xxxxxxxxxxx>).
>
> Edward
>
>
> On 9/1/06, Daniel Perez < daniel.perez@xxxxxxxxxxxxxxxx> wrote:
> > Hi Ed,
> >
> > I hope you have a couple of minutes to help me.
> >
> > I'm trying to install relax in Windows XP. First, I installed Python 2.4
> > with an interpreter for Windows named ActivePython. I also installed the
> > extensions Numeric_23.8 and ScientificPython-2.4.9.
> >
> > Then I got relax-1.2.6.src.tar.bz2, uncompressed it and copied the files
> > to my laptop.
> >
> > After adding the extension ".py" to the excecutable "relax", I could
> > sort of run it, well, I only got the python compiled versions for the
> > executables in the base directory (color, data, errors, io and processes
> > ".pyc") and that was all. No command window phython-looking-like or
> > anything similar.
> >
> > I tried to run "relax.py" from the Python IDE (using the "friendly"
> > ActivePython). The result was the same but I got some more information
> > from Python. The problem so far is that processes.py is trying to import
> > some linux commands (namely: kill, popopen3, and system)
> >
> > Here is the output from python:
> >
> > exec codeObject in __main__.__dict__
> > File "D:\Program Files\NMR\relax-1.2.6\relax.py", line 71, in ?
> > from generic_fns.main import Generic
> > File "D:\Program Files\NMR\relax-1.2.6\generic_fns\main.py", line 31,
> > in ?
> > from minimise import Minimise
> > File "D:\Program Files\NMR\relax-1.2.6\generic_fns\minimise.py", line
> > 27, in ?
> > from processes import RelaxPopen3
> > File "D:\Program Files\NMR\relax- 1.2.6\processes.py", line 23, in ?
> > from os import kill, popen3, system
> > ImportError: cannot import name kill
> >
> >
> >
> >
> >
> > I greped the scripts looking for more importing features and I got:
> >
> > processes.py: from os import kill, popen3, system
> >
> > relax: from os import F_OK,
access, getpid,
> > putenv
> >
> > io.py: from os import
devnull
> >
> > io.py: from os import F_OK,
access, makedirs,
> > remove, stat
> >
> > sconstruct: from os import F_OK,
access, chdir,
> > getcwd, lstat, path, remove, rmdir, sep, symlink, system, walk
> >
> > generic_fns/dasha.py: from os import F_OK,
access, chdir,
> > getcwd, system
> >
> > generic_fns/grace.py: from os import system
> >
> > generic_fns/molmol.py: from os import popen
> >
> > generic_fns/palmer.py: from os import F_OK,
P_WAIT, access,
> > chdir, chmod, getcwd, listdir, remove, spawnlp, system
> >
> > generic_fns/pdb.py: from os import F_OK,
access
> >
> > generic_fns/selection.py: from os import F_OK,
access
> >
> > prompt/interpreter.py: from os import F_OK,
access
> >
> > sample_scripts/full_analysis.py: from os import
getcwd, listdir
> >
> >
> >
> > For me it is kind of weird the phython did not complain about the
> > importing from os in the file relax (the second grep entry).
> >
> >
> > Do you think this could eventually work replacing these commands with
> > some windows stuff?
> >
> > many thanks in advance
> >
> > Daniel
> >
> >
> >
> > _______________________________________________
> > relax (http://nmr-relax.com )
> >
> > This is the relax-users mailing list
> > relax-users@xxxxxxx
> >
> > To unsubscribe from this list, get a password
> > reminder, or change your subscription options,
> > visit the list information page at
> > https://mail.gna.org/listinfo/relax-users
> >
>
> _______________________________________________
> relax ( http://nmr-relax.com)
>
> This is the relax-users mailing list
> relax-users@xxxxxxx
>
> To unsubscribe from this list, get a password
> reminder, or change your subscription options,
> visit the list information page at
> https://mail.gna.org/listinfo/relax-users
>
_______________________________________________
relax (http://nmr-relax.com)
This is the relax-users mailing list
relax-users@xxxxxxx
To unsubscribe from this list, get a password
reminder, or change your subscription options,
visit the list information page at
https://mail.gna.org/listinfo/relax-users