Ah, right, now I remember. If there are any remaining failures in the 1.3 line, it would be useful to have these all identified :) Especially if someone has PowerPC Mac hardware as well. I would like to release relax-1.3.5 soon and it would be good to have a 100% pass of the test-suite on both architectures. Do you know what the Scientific python issue Sebastian is seeing could be about? Cheers! Edward On 24 February 2010 15:30, Jack Howarth <howarth@xxxxxxxxxxxxxxxx> wrote:
Edward, I'll try to find some time to look at these issues over the weekend with the current 1.3 line code. As far as I recall, 1.3.4 still had some bogus testsuite failures on Darwin intel that were fixed for the next release of relax 1.3.x. Jack On Tue, Feb 23, 2010 at 02:57:23PM +0100, Edward d'Auvergne wrote:Hi, It might be worth talking with Jack Howarth (howarth att bromo dott med dott uc dott edu) as he is the maintainer of the relax fink packages. You can see our previous discussions about fink at: http://www.google.com/search?q=fink&btnG=Google+Search&domains=http%3A%2F%2Fnmr-relax.com&sitesearch=https%3A%2F%2Fmail.gna.org%2Fpublic%2Frelax-users He modified the sconstruct script for this, but I don't think this was merged upstream (it's not in the 1.3 line). Maybe all of these problems is because one of the other essential components of fink have been upgraded (numpy, python, etc.) causing a breakage in the relax C modules and in the Scientific python module. I you have the time and would like to try to get the 1.3 line test suite passing 100% and attempt to deposit the newest relax versions into fink, this would be appreciated. relax should run without problems on Mac OS X, and it shouldn't be too hard to modify relax for 100% compatibility. The only problem for me is that I don't have Mac hardware. Cheers, Edward On 22 February 2010 19:16, Sébastien Morin <sebastien.morin.1@xxxxxxxxx> wrote:Hi Ed, Unfortunately, the problem here arises from the fink version of relax, i.e. 1.3.4. Hence, the '-i' option is unavailable. For python, I indeed have two versions installed... 2.6: standard version 2.5: for rdiff-backup... For the sconstruct file, it was modified by fink. Hence, it seems the fink version of relax might be problematic... ... In order to see if the repository version is also having troubles, I installed it (as well as the minfx repository version) and tested it. The compilation with scons worked fine and relax does not complain anymore about the relax_fit.so file... However, there are still 12 errors from the unit tests and 1 failure and 16 errors from the system tests. It is better than with the fink version, although ScientificPython still not works... How should ScientificPython be installed ? Presently, it is installed from fink... ... Here is the output from 'relax -i': ========== Hardware information: Machine: i386 Processor: i386 System information: System: Darwin Release: 10.2.0 Version: Darwin Kernel Version 10.2.0: Tue Nov 3 10:37:10 PST 2009; root:xnu-1486.2.11~1/RELEASE_I386 Mac version: 10.6.2 (, , ) i386 Distribution: Full platform string: Darwin-10.2.0-i386-64bit Software information: Architecture: 64bit Python version: 2.6.1 Python branch: tags/r261 Python build: r261:67515, Jul 7 2009 23:51:51 Python compiler: GCC 4.2.1 (Apple Inc. build 5646) Python implementation: CPython Python revision: 67515 Numpy version: 1.2.1 Libc version: Network name: dhcp-160-189.rsvs.ulaval.ca ========== Any idea ? Séb :) On 10-02-22 12:41 PM, Edward d'Auvergne wrote:Hi Seb, Could you paste the output of 'relax -i'? This will help in working out what is happening and will be useful to have on record. Cheers, Edward On 22 February 2010 18:30, Edward d'Auvergne<edward@xxxxxxxxxxxxx> wrote:I just installed relax on a Mac using fink. relax-py26 1.3.4-2 Protein dynamics by NMR relax. data analysis I also installed the dependencies using fink. minfx-py26 1.0.2-1 The minfx optimisation library numpy-py26 [virtual package] python26 1:2.6.4-101 Interpreted, object-oriented language scientificpython-py26 2.8-1 Python modules for scientific computing scons 0.98.3-1 Python-based build tool I created a link in the relax directory pointing to the minfx directory: minfx@ -> /sw/lib/python2.6/site-packages/minfx/This should not be necessary. If there are any strange optimisation issues, this should be the first place to look.In fact, if the link is not present, relax complains about minfx not being installed...Ah, this is an indication that there are 2 python versions present!!! I have just added a new command line option to relax to help work out what is happening. I'm guessing you have a repository copy from the 1.3 line? If so, update this copy (for those who don't know, you type): $ svn up and then run one of: $ relax -i $ relax --info This should tell you all you need to know about versions, etc.! I can add other useful info later on - i.e. modules installed or missing and their versions.After, I launched relax but got the following warning: ======== ImportError: dlopen(/sw/lib/relax-py26/maths_fns/relax_fit.so, 2): no suitable image found. Did find: /sw/lib/relax-py26/maths_fns/relax_fit.so: mach-o, but wrong architecture Relaxation curve fitting is unavailable, try compiling the C modules. ========Try "scons clean_all" followed by "scons". Does the file /sw/lib/relax-py26/maths_fns/relax_fit.so exist? This will be a problem for curve-fitting!"scons clean_all" works fine and then, "scons" creates the "relax_fit.so" executable. The file type (obtained from the command "file") is as follows: "Mach-O bundle i386". Is it what it's supposed to be ? (I am really new to Mac...)I really have no idea either. Maybe someone else on this list knows something? It could actually be a version issue again as maybe relax is running with an old python version but scons build the module for the 2.6 version?Concerning the test-suite, there were nultiple errors: System tests 4 failures + 20 errors Unit tests 12 errors The unit tests errors were all the same: ====== Traceback (most recent call last): File "/sw/lib/relax-py26/test_suite/unit_tests/_generic_fns/_structure/test_scientific.py", line 54, in setUp self.data = Scientific_data() File "/sw/lib/relax-py26/generic_fns/structure/scientific.py", line 60, in __init__ raise RelaxError, "The Scientific python PDB module Scientific.IO.PDB could not be imported." RelaxError: RelaxError: The Scientific python PDB module Scientific.IO.PDB could not be imported. ====== ... I guess all these errors are not normal... Any idea to help me debug this issue ? Thank you !!!This is normal as Scientific python is not installed. Maybe it would be good to add a new section to the test suite listing the installed and missing modules, and then disabling the relevant system and unit tests (scientific-python, relax curve-fitting C modules, etc.) if missing. This will be a bit of work though, hence why I haven't done it yet! I would guess that all the failures are due to these 2 missing pieces, but it's hard to tell.Scientific Python is installed... scientificpython-py26 2.8-1 Python modules for scientific computing Maybe, yes, it would be good to simply give a warning for missing dependencies and avoid the tests to be performed... Thanks for your insights !This could be again related to Python versions, with Scientific Python installed for Python 2.6 but relax running with Python 1.5 (Mac Unix software versions are painfully old :P). I hope this helps. Regards, Edward-- Sébastien Morin PhD Student S. Gagné NMR Laboratory Université Laval& PROTEO Québec, Canada