started addressing issue #436

[unoebauer]

• Currently, only the runner is passed back to the model in the legacy mode
• Todo: set remaining packet diagnostic arrays
  • the runner has be accessible from inside the model
  • set remaining packet diagnostic arrays
  • model.to_hdf5 has to work
  • the tardis gui has to work
  • add tests

[wkerzendorf]

the diagnostic arrays are bound to the runner (at least they should be).

[unoebauer]

True - but to have full legacy suppport, the model should have some of these arrays (with the appropriate cuts) directly, see my comment below

[wkerzendorf]

thanks!

[unoebauer]

Looking at a simple tardis_example run, the model instance with is returned by run_tardis is still lacking:

• last_line_interaction_in_id
• last_line_interaction_out_id
• last_line_interaction_shell_id
• last_line_interaction_angstrom

This list is based on a quick inspection of include_from_model_in_hdf5 in model.py. One should also check the gui to make sure that nothing else is missing.

I'll take a look at that and will also add back the missing packet arrays.

[unoebauer]

After last commit, the to_hdf5 routine works again. Only irritating issue is the following warning that pops up:

mdl.to_hdf5("test.hdf5")
tardis.model - INFO - Writing to path 
py.warnings - WARNING - /home/ulrich/python-virtualenv/tardis/lib/python2.7/site-packages/pandas/io/pytables.py:2577: PerformanceWarning: 
your performance may suffer as PyTables will pickle object types that it cannot
map directly to c-types [inferred_type->mixed,key->values] [items->None]

  warnings.warn(ws, PerformanceWarning)

[unoebauer]

While running the gui, it emerged that the model is missing the following attributes:

• converged
• iteration_executed
• iteration_max_requested
• current_no_of_packets
• no_of_packets
• no_of_virtual_packets

[unoebauer]

With the latest commit, the gui is running again.

[unoebauer]

Added tests for tests for runner and model attributes. Worked offline - let's see if Travis is happy...

[unoebauer]

Can somebody have a proper look at this, @wkerzendorf? I'd say we are ready to merge