Expose settings root implicitly.

[prmukherj]

Settings root to be exposed implicitly.

For fluent versions till 23.1 there are a fixed set of settings attributes like: "file", "mesh", "setup", "solution", "results", "parametric_studies", "current_parametric_study", "parameters", "parallel", "report", and "server". For versions 23.2 and above, these attributes are directly taken from root and populated.

This removes hard coding of the attributes which were affected by changes to settings-api.

[raph-luc]

Question: The way we are providing dict(root) here is going to be missing attributes initially, compared to before. For example, with these changes I get:

>>> session = pf.launch_fluent()
>>> dir(session)
['__bool__', '__call__', '__class__', '__delattr__', '__dict__', '__dir__', '__doc__', '__enter__', '__eq__', '__exit__', '__format__', '__ge__', '__getattr__', '__getattribute__', '__getstate__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__le__', '__lt__', '__module__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', '__weakref__', '_batch_ops_service', '_build_from_fluent_connection', '_field_data_service', '_lck', '_monitors_service', '_pim_methods', '_preferences', '_reduction_service', '_remote_file_handler', '_root', '_se_service', '_settings_api_root', '_settings_root', '_settings_service', '_sync_from_future', '_system_coupling', '_tui', '_tui_service', '_version', '_workflow', '_workflow_se', 'build_from_fluent_connection', 'connection_properties', 'create_from_server_info_file', 'datamodel_events', 
'datamodel_service_se', 'datamodel_service_tui', 'download', 'error_state', 'events_manager', 'events_service', 'execute_tui', 'exit', 'field_data', 'field_data_streaming', 'field_info', 'fluent_connection', 'force_exit', 'force_exit_container', 'get_fluent_version', 'health_check_service', 'id', 'journal', 'monitors_manager', 'preferences', 'read_case', 'read_case_lightweight', 'reduction', 'rp_vars', 'scheme_eval', 'settings_service', 'start_journal', 'stop_journal', 'svar_data', 'svar_info', 'svar_service', 'system_coupling', 'transcript', 'tui', 'upload', 'version', 'workflow', 'write_case']

It is missing file, mesh, setup, solution, etc., regardless of existing codegen files (same behavior in v222 and v242). If I do any settings API call (which requests the static info), then after that point the dir() method includes the previously missing attributes/objects, as expected.

For comparison, in the main branch without these changes doing the same thing I get:

 >>> session = pf.launch_fluent()
>>> dir(session)
['__call__', '__class__', '__delattr__', '__dict__', '__dir__', '__doc__', '__enter__', '__eq__', '__exit__', '__format__', '__ge__', '__getattribute__', '__getstate__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__le__', '__lt__', '__module__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', '__weakref__', '_batch_ops_service', '_build_from_fluent_connection', '_field_data_service', '_lck', '_monitors_service', '_pim_methods', '_preferences', '_reduction_service', '_remote_file_handler', '_root', '_se_service', '_settings_root', '_settings_service', '_sync_from_future', '_system_coupling', '_tui', '_tui_service', '_version', '_workflow', '_workflow_se', 'build_from_fluent_connection', 'connection_properties', 'create_from_server_info_file', 'current_parametric_study', 'datamodel_events', 'datamodel_service_se', 'datamodel_service_tui', 'error_state', 'events_manager', 'events_service', 'execute_tui', 'exit', 'field_data', 'field_data_streaming', 'field_info', 'file', 'fluent_connection', 'force_exit', 'force_exit_container', 'get_fluent_version', 'health_check_service', 'id', 'journal', 'mesh', 'monitors_manager', 'parallel', 'parameters', 'parametric_studies', 'preferences', 'read_case', 'read_case_lightweight', 'reduction', 'report', 'results', 'rp_vars', 'scheme_eval', 'server', 'settings_service', 'setup', 'solution', 'start_journal', 'stop_journal', 'svar_data', 'svar_info', 'svar_service', 'system_coupling', 'transcript', 'tui', 'version', 'workflow', 'write_case']

Which as you can see has all of the missing attributes from the start.

Is this a known limitation of this approach?

Should we improve the __dir__() method so that it gets the attributes and properly populates the dir? As @mkundu1 suggested, I believe this could be done using the codegen files without needing to get the static info from the server.

[prmukherj]

Good point @raph-luc. I have updated the dir() with static info call. We are avoiding the use of codegen files for solver. Instead of that we are using static info call once to populate the list. Please point out if something else can be done. Thank you