Member ActiveStressODE::advance_time_step_local (const double t, const double dt, const double calcium, const double fiber_stretch, const double fiber_stretch_rate, Vector< double > &state) const override

[michelebucelli] It might be necessary or useful to implement other timestepping schemes, e.g. Runge-Kutta. In that case, we might want to expand the interface to support implicit time stepping too, e.g. by adding a method to evaluate the Jacobian matrix of the system, as in IonicModel.

Class ActiveStressRegazzoni

Force-strain-rate feedback requires a stabilization strategy for robust use in coupled electromechanics. This will be addressed in a follow-up PR.

Member BuenoOrovio::ModelParam

[TODO:DaveP] these guys should be maps map<int,double>.

[michelebucelli] Would it make sense to treat these in the same way as the epi/endo/myo zone are treated in the TTP model?

Member DomainParameters::stimuli

This uses unique_ptr unlike most similar containers because ParameterLists::params_map stores pointers to members of each object. Revisit this when the Parameters classes are refactored.

Member IonicModel::InitialStates

[michelebucelli] This would work better with a struct, due to the fields having meaningful names instead of first and second.

Namespace mat_fun

[TODO:DaveP] this should just be a namespace?

Member mat_fun::mat_inv_ge_orig (const Array< double > &A, const int nd, bool debug)

[TODO:DaveP] The original version sometimes produced NaNs.

Namespace read_files_ns

[TODO:DaveP] maxOutput=5 is is defined in consts but in the Fortran READEQ subroutine is defined as maxOutput=22.

Member read_files_ns::face_match (ComMod &com_mod, faceType &lFa, faceType &gFa, Vector< int > &ptr)

[TODO:DaveP] this has not been tested.

Member VtkData::get_point_data (const std::string &data_name) const

[michelebucelli] This should fall back onto copy_point_data.