Parse near_optimal results in SP

I’m debugging a SP model and the first solve GP solve returns this solution status

Interior-point solution summary
  Problem status  : PRIMAL_AND_DUAL_FEASIBLE
  Solution status : NEAR_OPTIMAL
  Primal.  obj: 1.1998763649e+01    Viol.  con: 3e-02    var: 0e+00  
  Dual.    obj: 1.2009877944e+01    Viol.  con: 0e+00    var: 7e-02  
* End solution on dual form. *

it looks like we currently don’t store any solution that isn’t both primal and dual feasible as well as optimal. I would like to access the near optimal solution and generate a table so I try and figure out what is going on.

I spoke to Woody and he had the idea that we should store all non-optimal solutions under an attribute like failed_solution…thoughts?

[bqpd]

dict(zip(m.program.gps[-1].varlocs, np.exp(m.program.gps[-1].solver_out["primal"])))

[mayork]

any updates on this? I think it would be very useful for debugging the near_feasible model