node_mapping / phase_mapping could be removed to simplify code

[EveCharbie]

No description provided.

[Ipuch]

Can you give a little overview of the code area expected to be touched ?

[Ipuch]

DONE !

[Ipuch]

I still have a bunch of questions about these functions @EveCharbie :

I think that passing the ocp was only necessary for phase mapping to work right ?

bioptim/bioptim/dynamics/configure_problem.py

Lines 121 to 138 in 39ed197

	def initialize(ocp, nlp):
	"""
	Call the dynamics a first time
	Parameters
	----------
	ocp: OptimalControlProgram
	A reference to the ocp
	nlp: NonLinearProgram
	A reference to the phase
	"""
	nlp.dynamics_type.type(
	ocp,
	nlp,
	numerical_data_timeseries=nlp.dynamics_type.numerical_data_timeseries,
	**nlp.dynamics_type.extra_parameters,
	)

Here I question the use of the function itself:

bioptim/bioptim/dynamics/configure_problem.py

Lines 1883 to 1919 in 39ed197

	def _apply_phase_mapping(ocp, nlp, name: str) -> BiMapping | None:
	"""
	Apply the phase mapping to the variable
	Parameters
	----------
	ocp: OptimalControlProgram
	A reference to the ocp
	nlp: NonLinearProgram
	A reference to the phase
	name: str
	The name of the variable to map
	Returns
	-------
	The mapping or None if no mapping is defined
	"""
	if nlp.phase_mapping:
	if name in nlp.variable_mappings.keys():
	double_mapping_to_first = (
	nlp.variable_mappings[name].to_first.map(nlp.phase_mapping.to_first.map_idx).T.tolist()[0]
	)
	double_mapping_to_first = [int(double_mapping_to_first[i]) for i in range(len(double_mapping_to_first))]
	double_mapping_to_second = (
	nlp.variable_mappings[name].to_second.map(nlp.phase_mapping.to_second.map_idx).T.tolist()[0]
	)
	double_mapping_to_second = [
	int(double_mapping_to_second[i]) for i in range(len(double_mapping_to_second))
	]
	else:
	double_mapping_to_first = nlp.phase_mapping.to_first.map_idx
	double_mapping_to_second = nlp.phase_mapping.to_second.map_idx
	axes_idx = BiMapping(to_first=double_mapping_to_first, to_second=double_mapping_to_second)
	else:
	axes_idx = None
	return axes_idx

That's only two right but there might have been some stuff hidden in the plot section right ?