Re: r6834 - /1.3/generic_fns/structure/geometric.py -- July 08, 2008 - 21:48

This could be caught in the system test, after the execution of the
script.  If the PDB is dumped into a temporary file, and then read
back in and checked if it has contents, that would be the best way to
catch this problem.  An example of where something like this already
happens can be found in
test_suite/unit_tests/sequence_testing_base.py.

Regards,

Edward


On Tue, Jul 8, 2008 at 8:23 PM, Sébastien Morin
<sebastien.morin.1@xxxxxxxxx> wrote:
> Hi Ed,
>
> The test passes when doing this change (since R is now defined).
>
> When using my script, I get a tensor pdb, but it is empty... since this
> is the same change as setting : R = centre_of_mass()...
>
> Any clue ?
> If the change of R to CoM is ok (within the create_diff_tensor_pdb()
> function), then maybe other changes are needed elsewhere...
>
> I'll commit these changes anyway. We can still revert later on if this
> is wrong.
>
> Regards,
>
>
> Séb
>
>
>
>
> Edward d'Auvergne wrote:
> > Sorry, I couldn't check the sources to find this as my computer is
> > having hardware failures at the moment.  But what you have found is
> > very likely to be the problem, and just changing all references to the
> > R variable to CoM should fix it.
> >
> > Cheers,
> >
> > Edward
> >
> >
> > On Tue, Jul 8, 2008 at 6:27 PM, Sébastien Morin
> > <sebastien.morin.1@xxxxxxxxx> wrote:
> >
> > > Hi Ed,
> > >
> > > I looked at the code using svn diff with revision number...
> > >
> > > It seems that line 225 of generic_fns/structure/geometric.py was changed
> > > from :
> > >
> > >    R = centre_of_mass()
> > >
> > > to
> > >
> > >    CoM = centre_of_mass()
> > >
> > > So, should we change the following calls to 'R' to calls to 'CoM' ? Or
> > > am I completely wrong..?
> > >
> > > Cheers,
> > >
> > >
> > > Séb
> > >
> > >
> > >
> > >
> > > Edward d'Auvergne wrote:
> > >
> > > > Hi,
> > > >
> > > > Yeah, I noticed that problem in your test.  I'll try to hunt down a
> > > > little later where we lost this R array (the origin set to the centre
> > > > of mass).  It should be documented in the changes in this method
> > > > (using svn diff with revision numbers will be useful) that have
> > > > occurred in the 1.3 line.
> > > >
> > > > Regards,
> > > >
> > > > Edward
> > > >
> > > >
> > > > On Tue, Jul 8, 2008 at 3:32 PM, Sébastien Morin
> > > > <sebastien.morin.1@xxxxxxxxx> wrote:
> > > >
> > > >
> > > > > Hi Ed,
> > > > >
> > > > > Everything seems cool ! I get to know the code much more, but may make
> > > > > some errors and introduce new bugs...
> > > > >
> > > > > Additionally, I a currently stuck with the message error for
> > > > > diffusion_tensor system test :
> > > > >
> > > > > ========
> > > > > line :
> > > > > Traceback (most recent call last):
> > > > >  File
> > > > > "/home/semor/pse-4/collaborations/relax/relax-1.3/test_suite/system_tests/diffusion_tensor.py",
> > > > > line 109, in test_createDiffTensor
> > > > >    
> > > > > self.relax.interpreter._Structure.create_diff_tensor_pdb(file='devnull')
> > > > >  File
> > > > > "/home/semor/pse-4/collaborations/relax/relax-1.3/prompt/structure.py",
> > > > > line 167, in create_diff_tensor_pdb
> > > > >    generic_fns.structure.geometric.create_diff_tensor_pdb(scale=scale,
> > > > > file=file, dir=dir, force=force)
> > > > >  File
> > > > > "/home/semor/pse-4/collaborations/relax/relax-1.3/generic_fns/structure/geometric.py",
> > > > > line 263, in create_diff_tensor_pdb
> > > > >    res_num = generate_vector_residues(structure=structure,
> > > > > vector=pipe.diff_tensor.Dpar*pipe.diff_tensor.Dpar_unit,
> > > > > atom_name='Dpar', res_name_vect='AXS', sim_vectors=sim_vectors,
> > > > > chain_id=chain_id, res_num=res_num, origin=R, scale=scale, neg=True)
> > > > > NameError: global name 'R' is not defined
> > > > > ========
> > > > >
> > > > > I tried (pessimistly) setting R to different things :
> > > > >
> > > > >    - pipe.diff_tensor.rotation
> > > > >    - centre_of_mass()
> > > > >    - eye(3)
> > > > >    - None
> > > > >
> > > > > but none would work (!) and I am now without idea... Do you have any
> > > > > clue ? Did I break something somewhere else yesterday ?
> > > > >
> > > > > Moreover, the system test doesn't stop at the first instance of
> > > > > structure.create_diff_tensor_pdb(), but at the second, the first being
> > > > > the one for the sphere. Hence, the sphere representation doesn't fail,
> > > > > but I tried it with a script of my own and it yields an empty pdb
> > > > > (except for the header). Would you have a tensor pdb I could look at so
> > > > > when the code works again, I know the file is correctly formatted..?
> > > > >
> > > > > Regards,
> > > > >
> > > > >
> > > > > Séb
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > Edward d'Auvergne wrote:
> > > > >
> > > > >
> > > > > > Hi,
> > > > > >
> > > > > > With the fixes you've been making and the changes I made to the
> > > > > > structural object code, I think the diffusion tensor representation
> > > > > > should soon be functioning again.  Do you see any real technical
> > > > > > issues preventing this?  I hope my changes haven't caused you too many
> > > > > > problems.  They were necessary though, as the PDB reading and writing
> > > > > > code was broken.
> > > > > >
> > > > > > Regards,
> > > > > >
> > > > > > Edward
> > > > > >
> > > > > >
> > > > > > On Tue, Jul 8, 2008 at 6:42 AM,  <sebastien.morin.1@xxxxxxxxx> wrote:
> > > > > >
> > > > > >
> > > > > >
> > > > > > > Author: semor
> > > > > > > Date: Tue Jul  8 06:42:46 2008
> > > > > > > New Revision: 6834
> > > > > > >
> > > > > > > URL: http://svn.gna.org/viewcvs/relax?rev=6834&view=rev
> > > > > > > Log:
> > > > > > > Corrected a typo introduced in r6815.
> > > > > > >
> > > > > > >
> > > > > > > Modified:
> > > > > > >    1.3/generic_fns/structure/geometric.py
> > > > > > >
> > > > > > > Modified: 1.3/generic_fns/structure/geometric.py
> > > > > > > URL: 
> > > > > > > http://svn.gna.org/viewcvs/relax/1.3/generic_fns/structure/geometric.py?rev=6834&r1=6833&r2=6834&view=diff
> > > > > > > ==============================================================================
> > > > > > > --- 1.3/generic_fns/structure/geometric.py (original)
> > > > > > > +++ 1.3/generic_fns/structure/geometric.py Tue Jul  8 06:42:46 2008
> > > > > > > @@ -611,7 +611,7 @@
> > > > > > >         atom_id_ext = ''
> > > > > > >
> > > > > > >     # The origin atom.
> > > > > > > -    structure.atom_add(pdb_record='HETATM', 
> > > > > > > atom_num='R_vect'+atom_id_ext, atom_name=R, res_name=res_name_vect, 
> > > > > > > chain_id=chain_id, res_num=res_num, pos=origin, segment_id=None, 
> > > > > > > element='C', struct_index=None)
> > > > > > > +    structure.atom_add(pdb_record='HETATM', 
> > > > > > > atom_num='R_vect'+atom_id_ext, atom_name='R', res_name=res_name_vect, 
> > > > > > > chain_id=chain_id, res_num=res_num, pos=origin, segment_id=None, 
> > > > > > > element='C', struct_index=None)
> > > > > > >
> > > > > > >     # Create the PDB residue representing the vector.
> > > > > > >     structure.atom_add(pdb_record='HETATM', 
> > > > > > > atom_num=atom_name+atom_id_ext, atom_name=atom_name, 
> > > > > > > res_name=res_name_vect, chain_id=chain_id, res_num=res_num, 
> > > > > > > pos=origin+vector*scale, segment_id=None, element='C', 
> > > > > > > struct_index=None)
> > > > > > >
> > > > > > >
> > > > > > > _______________________________________________
> > > > > > > relax (http://nmr-relax.com)
> > > > > > >
> > > > > > > This is the relax-commits mailing list
> > > > > > > relax-commits@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-commits
> > > > > > _______________________________________________
> > > > > > relax (http://nmr-relax.com)
> > > > > >
> > > > > > This is the relax-devel mailing list
> > > > > > relax-devel@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-devel