Re: value.set function create the err version of the parameter. -- October 24, 2013 - 08:50

Hi,

It's not a problem.  This user function is not ideally designed yet.
Maybe changing the argument from 'error' to 'set_error' would improve
it.  This user function also needs to be modified to be able to set
list and dictionary type parameter values, but this is very difficult.
 For example the r2 parameter in the dispersion analysis which is used
for the R2eff or R1rho rate in the absence of exchange.  This is
currently a list with the elements ordered with the order of
spectrometer frequencies.  But you cannot assume this structure as it
can change in the future (as often happens).  And in the user
functions, the aim is to avoid all programming concepts so the user
should not be presented with the concept of lists and dictionaries, or
at least this should be minimised.  So there has to be some
abstraction for the user - i.e. they simply specify the value and the
qualifier which is the magnetic field strength.  Anyway, that was a
bit of an aside!

Regards,

Edward



On 23 October 2013 21:28, Troels Emtekær Linnet <tlinnet@xxxxxxxxxxxxx> wrote:
> Oh, well, that is embarrasing.
>
> I understood from the help lines, that the error flag was to return error
> messages, if the value was not set.
>
> Silly.
>
> Den 23/10/2013 18.50 skrev "Edward d'Auvergne" <edward@xxxxxxxxxxxxx>:
>
> > Hi,
> >
> > Was this unexpected?  The error flag allows the parameter error rather
> > than value to be set, so that is the correct behaviour.  Were you
> > trying to do something different?
> >
> > Regards,
> >
> > Edward
> >
> >
> >
> >
> > On 23 October 2013 18:26, Troels Emtekær Linnet <tlinnet@xxxxxxxxxxxxx>
> > wrote:
> > > Hi Edward.
> > >
> > > This test script:
> > >
> > > ----------
> > > pipe.create(pipe_name='test', bundle='test', pipe_type='relax_disp')
> > >
> > > spin.create(spin_name='N', spin_num=1, res_name='V', res_num=2,
> > > mol_name=None)
> > > spin.create(spin_name='N', spin_num=2, res_name='F', res_num=3,
> > > mol_name=None)
> > > spin.create(spin_name='N', spin_num=3, res_name='G', res_num=4,
> > > mol_name=None)
> > > spin.create(spin_name='N', spin_num=4, res_name='R', res_num=5,
> > > mol_name=None)
> > > spin.create(spin_name='N', spin_num=5, res_name='C', res_num=6,
> > > mol_name=None)
> > >
> > > relax_disp.cluster('model_cluster', ':2@N')
> > > relax_disp.cluster('model_cluster', ':3@N')
> > >
> > > value.set(val = 8.5, param='r2', spin_id=':2@N', error=True)
> > > value.set(val = 500, param='kex', spin_id=':3@N', error=True)
> > >
> > > state.save('test', force=True)
> > > ----------
> > >
> > > If I load the saved state, and inspect the spins in the spin viewer, I
> > > see
> > > that the
> > > parameters are set for:
> > > r2_err = 8.5
> > > kex_err = 500
> > >
> > > Is this the expected behavior?
> > >
> > > best
> > > Troels
> > >
> > > _______________________________________________
> > > relax (http://www.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