Fix for bug #20417 (https://gna.org/bugs/?20417).

The backwards compatibility hook for the loading of old relax state and results files can now handle
the presence of proton spins. The data structures are now correctly converted to the current data
pipe design. This is only the case for old files containing PCS data.

(Browse SVN revision 18182)

It appears that the backwards compatibility hooks for loading old state files fails for the combination of a state with alignment data, spin data, and Monte Carlo simulations. This is now caught by the State.test_align_tensor_with_mc_sims system test. The error message from the test (or running relax with the debug flag) is:

relax> state.load(state='/data/edau/relax/relax-trunk/test_suite/shared_data/saved_states/align_tensor_mc.bz2', dir=None, force=False)
Opening the file '/data/edau/relax/relax-trunk/test_suite/shared_data/saved_states/align_tensor_mc.bz2' for reading.

Traceback (most recent call last):
File "test_suite/system_tests/state.py", line 70, in test_align_tensor_with_mc_sims
self.interpreter.state.load(path)
File "prompt/uf_objects.py", line 219, in _call_
self._backend(new_args, *uf_kargs)
File "generic_fns/state.py", line 101, in load_state
ds.from_xml(file)
File "data/__init__.py", line 489, in from_xml
self._back_compat_hook(file_version, pipes=pipes)
File "data/__init__.py", line 240, in _back_compat_hook
h_spin = generic_fns.mol_res_spin.create_spin(mol_name=mol.name, res_num=res.num, res_name=res.name, spin_name=proton_name, pipe=pipe_name)
File "generic_fns/mol_res_spin.py", line 1224, in create_spin
res_cont.spin.add_item(spin_num=spin_num, spin_name=spin_name)
File "data/mol_res_spin.py", line 258, in add_item
raise RelaxError("The unnumbered spin name '%s' already exists." % spin_name)
RelaxError: RelaxError: The unnumbered spin name 'H' already exists.