summaryrefslogtreecommitdiff
path: root/sci_gateway/cpp
diff options
context:
space:
mode:
Diffstat (limited to 'sci_gateway/cpp')
-rw-r--r--sci_gateway/cpp/QuadTMINLP.hpp134
-rw-r--r--sci_gateway/cpp/bonmin.opt156
-rwxr-xr-xsci_gateway/cpp/builder_gateway_cpp.sce68
-rw-r--r--sci_gateway/cpp/cbcintlinprog_matrixcpp.cpp226
-rw-r--r--sci_gateway/cpp/cbcintlinprog_mpscpp.cpp115
-rwxr-xr-xsci_gateway/cpp/cleaner.sce22
-rw-r--r--sci_gateway/cpp/cpp_intfminbnd.cpp172
-rw-r--r--sci_gateway/cpp/cpp_intfmincon.cpp189
-rw-r--r--sci_gateway/cpp/cpp_intfminunc.cpp174
-rw-r--r--sci_gateway/cpp/cpp_intqpipopt.cpp267
-rw-r--r--sci_gateway/cpp/libFOSSEE_Optimization_Toolbox.c40
-rwxr-xr-xsci_gateway/cpp/libFOSSEE_Optimization_Toolbox.sobin0 -> 129207 bytes
-rw-r--r--sci_gateway/cpp/loader.sce26
-rw-r--r--sci_gateway/cpp/minbndTMINLP.hpp114
-rw-r--r--sci_gateway/cpp/minconTMINLP.hpp124
-rw-r--r--sci_gateway/cpp/minuncTMINLP.hpp113
-rw-r--r--sci_gateway/cpp/sci_QuadTMINLP.cpp230
-rw-r--r--sci_gateway/cpp/sci_iofunc.cpp333
-rw-r--r--sci_gateway/cpp/sci_iofunc.hpp25
-rw-r--r--sci_gateway/cpp/sci_minbndTMINLP.cpp218
-rw-r--r--sci_gateway/cpp/sci_minconTMINLP.cpp324
-rw-r--r--sci_gateway/cpp/sci_minuncTMINLP.cpp236
22 files changed, 3306 insertions, 0 deletions
diff --git a/sci_gateway/cpp/QuadTMINLP.hpp b/sci_gateway/cpp/QuadTMINLP.hpp
new file mode 100644
index 0000000..84704be
--- /dev/null
+++ b/sci_gateway/cpp/QuadTMINLP.hpp
@@ -0,0 +1,134 @@
+// Copyright (C) 2016 - IIT Bombay - FOSSEE
+//
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+// Author: Harpreet Singh, Pranav Deshpande and Akshay Miterani
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+
+#ifndef QuadTMINLP_HPP
+#define QuadTMINLP_HPP
+
+#include "BonTMINLP.hpp"
+#include "IpTNLP.hpp"
+
+using namespace Ipopt;
+using namespace Bonmin;
+
+class QuadTMINLP : public TMINLP
+{
+ private:
+ Index numVars_; // Number of variables.
+
+ Index numCons_; // Number of constraints.
+
+ Index intconSize_; // Number of integer constraints
+
+ const Number *qMatrix_ = NULL; //qMatrix_ is a pointer to matrix of size numVars X numVars_
+ // with coefficents of quadratic terms in objective function.
+
+ const Number *lMatrix_ = NULL;//lMatrix_ is a pointer to matrix of size 1*numVars_
+ // with coefficents of linear terms in objective function.
+
+ const Number *intcon_ = NULL; // The matrix containing the integer constraints
+
+ const Number *conMatrix_ = NULL;//conMatrix_ is a pointer to matrix of size numCons X numVars
+ // with coefficients of terms in a each objective in each row.
+
+ const Number *conUB_= NULL; //conUB_ is a pointer to a matrix of size of 1*numCons_
+ // with upper bounds of all constraints.
+
+ const Number *conLB_ = NULL; //conLB_ is a pointer to a matrix of size of 1*numConsn_
+ // with lower bounds of all constraints.
+
+ const Number *varUB_= NULL; //varUB_ is a pointer to a matrix of size of 1*numVar_
+ // with upper bounds of all variables.
+
+ const Number *varLB_= NULL; //varLB_ is a pointer to a matrix of size of 1*numVar_
+ // with lower bounds of all variables.
+
+ const Number *varGuess_= NULL; //varGuess_ is a pointer to a matrix of size of 1*numVar_
+ // with initial guess of all variables.
+
+ Number *finalX_= NULL; //finalX_ is a pointer to a matrix of size of 1*numVar_
+ // with final value for the primal variables.
+
+ Number *finalZl_= NULL; //finalZl_ is a pointer to a matrix of size of 1*numVar_
+ // with final values for the lower bound multipliers
+
+ Number *finalZu_= NULL; //finalZu_ is a pointer to a matrix of size of 1*numVar_
+ // with final values for the upper bound multipliers
+
+ Number *finalLambda_= NULL; //finalLambda_ is a pointer to a matrix of size of 1*numConstr_
+ // with final values for the upper bound multipliers
+
+ Number finalObjVal_; //finalObjVal_ is a scalar with the final value of the objective.
+
+ int status_; //Solver return status
+
+public:
+ // Constructor
+ QuadTMINLP(Index nV, Index nC, Index intconSize,Number *qM, Number *lM, Number *intcon,Number *cM, Number *cLB, Number *cUB, Number *vLB, Number *vUB,Number *vG):
+ numVars_(nV),numCons_(nC),intconSize_(intconSize),qMatrix_(qM),lMatrix_(lM),intcon_(intcon),conMatrix_(cM),conLB_(cLB),conUB_(cUB),varLB_(vLB),varUB_(vUB),varGuess_(vG), finalObjVal_(0){ }
+
+ // virtual destructor.
+ virtual ~QuadTMINLP(){}
+
+ /* Copy constructor.*/
+ QuadTMINLP(const QuadTMINLP &other){}
+
+ // Go to http://coin-or.org/Bonmin for the details of the below methods
+
+ virtual bool get_variables_types(Index n, VariableType* var_types);
+
+ virtual bool get_variables_linearity(Index n, Ipopt::TNLP::LinearityType* var_types);
+
+ virtual bool get_constraints_linearity(Index m, Ipopt::TNLP::LinearityType* const_types);
+
+
+ virtual bool get_nlp_info(Index& n, Index&m, Index& nnz_jac_g,
+ Index& nnz_h_lag, TNLP::IndexStyleEnum& index_style);
+
+ virtual bool get_bounds_info(Index n, Number* x_l, Number* x_u,
+ Index m, Number* g_l, Number* g_u);
+
+ virtual bool get_starting_point(Index n, bool init_x, Number* x,
+ bool init_z, Number* z_L, Number* z_U,
+ Index m, bool init_lambda,
+ Number* lambda);
+
+ virtual bool eval_f(Index n, const Number* x, bool new_x, Number& obj_value);
+
+ virtual bool eval_grad_f(Index n, const Number* x, bool new_x, Number* grad_f);
+
+ virtual bool eval_g(Index n, const Number* x, bool new_x, Index m, Number* g);
+
+ virtual bool eval_jac_g(Index n, const Number* x, bool new_x,
+ Index m, Index nele_jac, Index* iRow, Index *jCol,
+ Number* values);
+
+ virtual bool eval_h(Index n, const Number* x, bool new_x,
+ Number obj_factor, Index m, const Number* lambda,
+ bool new_lambda, Index nele_hess, Index* iRow,
+ Index* jCol, Number* values);
+
+ virtual void QuadTMINLP::finalize_solution(TMINLP::SolverReturn status,
+ Index n, const Number* x,Number obj_value);
+
+ virtual const SosInfo * sosConstraints() const{return NULL;}
+ virtual const BranchingInfo* branchingInfo() const{return NULL;}
+
+ // Bonmin methods end here
+
+ virtual const double * getX(); //Returns a pointer to a matrix of size of 1*numVar
+ // with final value for the primal variables.
+
+ virtual double getObjVal(); //Returns the output of the final value of the objective.
+
+ virtual int returnStatus(); //Returns the status count
+};
+
+#endif
diff --git a/sci_gateway/cpp/bonmin.opt b/sci_gateway/cpp/bonmin.opt
new file mode 100644
index 0000000..72fc254
--- /dev/null
+++ b/sci_gateway/cpp/bonmin.opt
@@ -0,0 +1,156 @@
+
+# registering category: Algorithm choice
+
+bonmin.algorithm B-BB #Choice of the algorithm.
+
+# registering category: Branch-and-bound options
+
+bonmin.allowable_fraction_gap 0 #Specify the value of relative gap under which the algorithm stops.
+bonmin.allowable_gap 0 #Specify the value of absolute gap under which the algorithm stops.
+bonmin.cutoff 1e+100 #Specify cutoff value.
+bonmin.cutoff_decr 1e-05 #Specify cutoff decrement.
+bonmin.enable_dynamic_nlp no #Enable dynamic linear and quadratic rows addition in nlp
+bonmin.integer_tolerance 1e-06 #Set integer tolerance.
+bonmin.iteration_limit 2147483647 #Set the cumulative maximum number of iteration in the algorithm used to process nodes continuous relaxations in the branch-and-bound.
+bonmin.nlp_failure_behavior stop #Set the behavior when an NLP or a series of NLP are unsolved by Ipopt (we call unsolved an NLP for which Ipopt is not able to guarantee optimality within the specified tolerances).
+bonmin.node_comparison best-bound #Choose the node selection strategy.
+bonmin.node_limit 2147483647 #Set the maximum number of nodes explored in the branch-and-bound search.
+bonmin.num_cut_passes 1 #Set the maximum number of cut passes at regular nodes of the branch-and-cut.
+bonmin.num_cut_passes_at_root 20 #Set the maximum number of cut passes at regular nodes of the branch-and-cut.
+bonmin.number_before_trust 8 #Set the number of branches on a variable before its pseudo costs are to be believed in dynamic strong branching.
+bonmin.number_strong_branch 20 #Choose the maximum number of variables considered for strong branching.
+bonmin.random_generator_seed 0 #Set seed for random number generator (a value of -1 sets seeds to time since Epoch).
+bonmin.read_solution_file no #Read a file with the optimal solution to test if algorithms cuts it.
+bonmin.solution_limit 2147483647 #Abort after that much integer feasible solution have been found by algorithm
+bonmin.sos_constraints enable #Whether or not to activate SOS constraints.
+bonmin.time_limit 1e+10 #Set the global maximum computation time (in secs) for the algorithm.
+bonmin.tree_search_strategy probed-dive #Pick a strategy for traversing the tree
+bonmin.variable_selection strong-branching #Chooses variable selection strategy
+
+# registering category: ECP cuts generation
+
+bonmin.ecp_abs_tol 1e-06 #Set the absolute termination tolerance for ECP rounds.
+bonmin.ecp_max_rounds 5 #Set the maximal number of rounds of ECP cuts.
+bonmin.ecp_probability_factor 10 #Factor appearing in formula for skipping ECP cuts.
+bonmin.ecp_rel_tol 0 #Set the relative termination tolerance for ECP rounds.
+bonmin.filmint_ecp_cuts 0 #Specify the frequency (in terms of nodes) at which some a la filmint ecp cuts are generated.
+
+# registering category: Feasibility checker using OA cuts
+
+bonmin.feas_check_cut_types outer-approx #Choose the type of cuts generated when an integer feasible solution is found
+bonmin.feas_check_discard_policy detect-cycles #How cuts from feasibility checker are discarded
+bonmin.generate_benders_after_so_many_oa 5000 #Specify that after so many oa cuts have been generated Benders cuts should be generated instead.
+
+# registering category: MILP Solver
+
+bonmin.cpx_parallel_strategy 0 #Strategy of parallel search mode in CPLEX.
+bonmin.milp_solver Cbc_D #Choose the subsolver to solve MILP sub-problems in OA decompositions.
+bonmin.milp_strategy solve_to_optimality #Choose a strategy for MILPs.
+bonmin.number_cpx_threads 0 #Set number of threads to use with cplex.
+
+# registering category: MILP cutting planes in hybrid algorithm
+
+bonmin.2mir_cuts 0 #Frequency (in terms of nodes) for generating 2-MIR cuts in branch-and-cut
+bonmin.Gomory_cuts -5 #Frequency (in terms of nodes) for generating Gomory cuts in branch-and-cut.
+bonmin.clique_cuts -5 #Frequency (in terms of nodes) for generating clique cuts in branch-and-cut
+bonmin.cover_cuts 0 #Frequency (in terms of nodes) for generating cover cuts in branch-and-cut
+bonmin.flow_cover_cuts -5 #Frequency (in terms of nodes) for generating flow cover cuts in branch-and-cut
+bonmin.lift_and_project_cuts 0 #Frequency (in terms of nodes) for generating lift-and-project cuts in branch-and-cut
+bonmin.mir_cuts -5 #Frequency (in terms of nodes) for generating MIR cuts in branch-and-cut
+bonmin.reduce_and_split_cuts 0 #Frequency (in terms of nodes) for generating reduce-and-split cuts in branch-and-cut
+
+# registering category: NLP interface
+
+bonmin.nlp_solver Ipopt #Choice of the solver for local optima of continuous NLP's
+bonmin.warm_start none #Select the warm start method
+
+# registering category: NLP solution robustness
+
+bonmin.max_consecutive_failures 10 #(temporarily removed) Number $n$ of consecutive unsolved problems before aborting a branch of the tree.
+bonmin.max_random_point_radius 100000 #Set max value r for coordinate of a random point.
+bonmin.num_iterations_suspect -1 #Number of iterations over which a node is considered "suspect" (for debugging purposes only, see detailed documentation).
+bonmin.num_retry_unsolved_random_point 0 #Number $k$ of times that the algorithm will try to resolve an unsolved NLP with a random starting point (we call unsolved an NLP for which Ipopt is not able to guarantee optimality within the specified tolerances).
+bonmin.random_point_perturbation_interval 1 #Amount by which starting point is perturbed when choosing to pick random point by perturbing starting point
+bonmin.random_point_type Jon #method to choose a random starting point
+bonmin.resolve_on_small_infeasibility 0 #If a locally infeasible problem is infeasible by less than this, resolve it with initial starting point.
+
+# registering category: NLP solves in hybrid algorithm (B-Hyb)
+
+bonmin.nlp_solve_frequency 10 #Specify the frequency (in terms of nodes) at which NLP relaxations are solved in B-Hyb.
+bonmin.nlp_solve_max_depth 10 #Set maximum depth in the tree at which NLP relaxations are solved in B-Hyb.
+bonmin.nlp_solves_per_depth 1e+100 #Set average number of nodes in the tree at which NLP relaxations are solved in B-Hyb for each depth.
+
+# registering category: Nonconvex problems
+
+bonmin.coeff_var_threshold 0.1 #Coefficient of variation threshold (for dynamic definition of cutoff_decr).
+bonmin.dynamic_def_cutoff_decr no #Do you want to define the parameter cutoff_decr dynamically?
+bonmin.first_perc_for_cutoff_decr -0.02 #The percentage used when, the coeff of variance is smaller than the threshold, to compute the cutoff_decr dynamically.
+bonmin.max_consecutive_infeasible 0 #Number of consecutive infeasible subproblems before aborting a branch.
+bonmin.num_resolve_at_infeasibles 0 #Number $k$ of tries to resolve an infeasible node (other than the root) of the tree with different starting point.
+bonmin.num_resolve_at_node 0 #Number $k$ of tries to resolve a node (other than the root) of the tree with different starting point.
+bonmin.num_resolve_at_root 0 #Number $k$ of tries to resolve the root node with different starting points.
+bonmin.second_perc_for_cutoff_decr -0.05 #The percentage used when, the coeff of variance is greater than the threshold, to compute the cutoff_decr dynamically.
+
+# registering category: Outer Approximation Decomposition (B-OA)
+
+bonmin.oa_decomposition no #If yes do initial OA decomposition
+
+# registering category: Outer Approximation cuts generation
+
+bonmin.add_only_violated_oa no #Do we add all OA cuts or only the ones violated by current point?
+bonmin.oa_cuts_scope global #Specify if OA cuts added are to be set globally or locally valid
+bonmin.oa_rhs_relax 1e-08 #Value by which to relax OA cut
+bonmin.tiny_element 1e-08 #Value for tiny element in OA cut
+bonmin.very_tiny_element 1e-17 #Value for very tiny element in OA cut
+
+# registering category: Output
+
+bonmin.bb_log_interval 100 #Interval at which node level output is printed.
+bonmin.bb_log_level 1 #specify main branch-and-bound log level.
+bonmin.file_print_level 5 #Verbosity level for output file.
+bonmin.file_solution no #Write a file bonmin.sol with the solution
+bonmin.fp_log_frequency 100 #display an update on lower and upper bounds in FP every n seconds
+bonmin.fp_log_level 1 #specify FP iterations log level.
+bonmin.inf_pr_output original #Determines what value is printed in the "inf_pr" output column.
+bonmin.lp_log_level 0 #specify LP log level.
+bonmin.milp_log_level 0 #specify MILP solver log level.
+bonmin.nlp_log_at_root 0 # Specify a different log level for root relaxation.
+bonmin.nlp_log_level 1 #specify NLP solver interface log level (independent from ipopt print_level).
+bonmin.oa_cuts_log_level 0 #level of log when generating OA cuts.
+bonmin.oa_log_frequency 100 #display an update on lower and upper bounds in OA every n seconds
+bonmin.oa_log_level 1 #specify OA iterations log level.
+bonmin.option_file_name #File name of options file (to overwrite default).
+bonmin.output_file #File name of desired output file (leave unset for no file output).
+bonmin.print_frequency_iter 1 #Determines at which iteration frequency the summarizing iteration output line should be printed.
+bonmin.print_frequency_time 0 #Determines at which time frequency the summarizing iteration output line should be printed.
+bonmin.print_info_string no #Enables printing of additional info string at end of iteration output.
+bonmin.print_level 5 #Output verbosity level.
+bonmin.print_options_documentation no #Switch to print all algorithmic options.
+bonmin.print_timing_statistics no #Switch to print timing statistics.
+bonmin.print_user_options no #Print all options set by the user.
+bonmin.replace_bounds no #Indicates if all variable bounds should be replaced by inequality constraints
+bonmin.skip_finalize_solution_call no #Indicates if call to NLP::FinalizeSolution after optimization should be suppressed
+
+# registering category: Primal Heuristics
+
+bonmin.feasibility_pump_objective_norm 1 #Norm of feasibility pump objective function
+bonmin.fp_pass_infeasible no #Say whether feasibility pump should claim to converge or not
+bonmin.heuristic_RINS no #if yes runs the RINS heuristic
+bonmin.heuristic_dive_MIP_fractional no #if yes runs the Dive MIP Fractional heuristic
+bonmin.heuristic_dive_MIP_vectorLength no #if yes runs the Dive MIP VectorLength heuristic
+bonmin.heuristic_dive_fractional no #if yes runs the Dive Fractional heuristic
+bonmin.heuristic_dive_vectorLength no #if yes runs the Dive VectorLength heuristic
+bonmin.heuristic_feasibility_pump no #whether the heuristic feasibility pump should be used
+bonmin.pump_for_minlp no #whether to run the feasibility pump heuristic for MINLP
+
+# registering category: Strong branching setup
+
+bonmin.candidate_sort_criterion best-ps-cost #Choice of the criterion to choose candidates in strong-branching
+bonmin.maxmin_crit_have_sol 0.1 #Weight towards minimum in of lower and upper branching estimates when a solution has been found.
+bonmin.maxmin_crit_no_sol 0.7 #Weight towards minimum in of lower and upper branching estimates when no solution has been found yet.
+bonmin.min_number_strong_branch 0 #Sets minimum number of variables for strong branching (overriding trust)
+bonmin.number_before_trust_list 0 #Set the number of branches on a variable before its pseudo costs are to be believed during setup of strong branching candidate list.
+bonmin.number_look_ahead 0 #Sets limit of look-ahead strong-branching trials
+bonmin.number_strong_branch_root 2147483647 #Maximum number of variables considered for strong branching in root node.
+bonmin.setup_pseudo_frac 0.5 #Proportion of strong branching list that has to be taken from most-integer-infeasible list.
+bonmin.trust_strong_branching_for_pseudo_cost yes #Whether or not to trust strong branching results for updating pseudo costs.
diff --git a/sci_gateway/cpp/builder_gateway_cpp.sce b/sci_gateway/cpp/builder_gateway_cpp.sce
new file mode 100755
index 0000000..1359db8
--- /dev/null
+++ b/sci_gateway/cpp/builder_gateway_cpp.sce
@@ -0,0 +1,68 @@
+// Copyright (C) 2015 - IIT Bombay - FOSSEE
+//
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+// Author: Harpreet Singh
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+
+mode(-1)
+lines(0)
+
+toolbox_title = "FOSSEE_Optimization_Toolbox";
+
+Build_64Bits = %t;
+
+path_builder = get_absolute_file_path('builder_gateway_cpp.sce');
+
+//Name of All the Functions
+Function_Names = [
+ 'inter_fminunc', 'cpp_intfminunc';
+ 'inter_fminbnd', 'cpp_intfminbnd';
+ 'inter_fmincon', 'cpp_intfmincon';
+ 'sci_intqpipopt', 'cpp_intqpipopt';
+ 'sci_matrix_intlinprog', 'matrix_cppintlinprog';
+ 'sci_mps_intlinprog','mps_cppintlinprog';
+ ];
+
+//Name of all the files to be compiled
+Files = [
+ 'sci_iofunc.cpp',
+ 'sci_minuncTMINLP.cpp',
+ 'cpp_intfminunc.cpp',
+ 'sci_minbndTMINLP.cpp',
+ 'cpp_intfminbnd.cpp',
+ 'sci_minconTMINLP.cpp',
+ 'cpp_intfmincon.cpp',
+ 'cbcintlinprog_matrixcpp.cpp',
+ 'sci_QuadTMINLP.cpp',
+ 'cpp_intqpipopt.cpp',
+ 'cbcintlinprog_mpscpp.cpp'
+ ]
+
+[a, opt] = getversion();
+Version = opt(2);
+
+if getos()=="Windows" then
+ third_dir = path_builder+filesep()+'..'+filesep()+'..'+filesep()+'thirdparty';
+ lib_base_dir = third_dir + filesep() + 'windows' + filesep() + 'lib' + filesep() + Version + filesep();
+ inc_base_dir = third_dir + filesep() + 'windows' + filesep() + 'include' + filesep() + 'coin';
+// C_Flags=['-D__USE_DEPRECATED_STACK_FUNCTIONS__ -w -I '+path_builder+' '+ '-I '+inc_base_dir+' ']
+// Linker_Flag = [lib_base_dir+"libClp.lib "+lib_base_dir+"libCgl.lib "+lib_base_dir+"libOsi.lib "+lib_base_dir+"libOsiClp.lib "+lib_base_dir+"libCoinUtils.lib "+lib_base_dir+"libSymphony.lib "+lib_base_dir+"IpOptFSS.lib "+lib_base_dir+"IpOpt-vc10.lib "]
+
+else
+ third_dir = path_builder+filesep()+'..'+filesep()+'..'+filesep()+'thirdparty';
+ lib_base_dir = third_dir + filesep() + 'linux' + filesep() + 'lib' + filesep() + Version + filesep();
+ inc_base_dir = third_dir + filesep() + 'linux' + filesep() + 'include' + filesep() + 'coin';
+
+ C_Flags=["-D__USE_DEPRECATED_STACK_FUNCTIONS__ -w -fpermissive -I"+path_builder+" -I"+inc_base_dir+" -Wl,-rpath="+lib_base_dir+" "]
+
+ Linker_Flag = ["-L"+lib_base_dir+"libCoinUtils.so "+"-L"+lib_base_dir+"libClp.so "+"-L"+lib_base_dir+"libClpSolver.so "+"-L"+lib_base_dir+"libOsi.so "+"-L"+lib_base_dir+"libOsiClp.so "+"-L"+lib_base_dir+"libCgl.so "+"-L"+lib_base_dir+"libCbc.so "+"-L"+lib_base_dir+"libCbcSolver.so "+"-L"+lib_base_dir+"libOsiCbc.so "+"-L"+lib_base_dir+"libipopt.so "+"-L"+lib_base_dir+"libbonmin.so " ]
+end
+
+tbx_build_gateway(toolbox_title,Function_Names,Files,get_absolute_file_path("builder_gateway_cpp.sce"), [], Linker_Flag, C_Flags);
+
+clear toolbox_title Function_Names Files Linker_Flag C_Flags;
diff --git a/sci_gateway/cpp/cbcintlinprog_matrixcpp.cpp b/sci_gateway/cpp/cbcintlinprog_matrixcpp.cpp
new file mode 100644
index 0000000..d4e6f41
--- /dev/null
+++ b/sci_gateway/cpp/cbcintlinprog_matrixcpp.cpp
@@ -0,0 +1,226 @@
+// MILP with CBC library, Matrix
+// Code Authors: Akshay Miterani and Pranav Deshpande
+
+#include <sci_iofunc.hpp>
+
+// For Branch and bound
+#include "OsiSolverInterface.hpp"
+#include "CbcModel.hpp"
+#include "CbcCutGenerator.hpp"
+#include "CbcHeuristicLocal.hpp"
+#include "OsiClpSolverInterface.hpp"
+extern "C"{
+#include <api_scilab.h>
+#include "sciprint.h"
+
+int matrix_cppintlinprog(){
+
+ //Objective function
+ double* obj;
+ //Constraint matrix coefficients
+ double* conMatrix;
+ //intcon Matrix
+ double* intcon;
+ //Constraints upper bound
+ double* conlb;
+ //Constraints lower bound
+ double* conub;
+ //Lower bounds for variables
+ double* lb;
+ //Upper bounds for variables
+ double* ub;
+ //options for maximum iterations and writing mps
+ double* options;
+ //Flag for Mps
+ double flagMps;
+ //mps file path
+ char * mpsFile;
+ //Error structure in Scilab
+ SciErr sciErr;
+ //Number of rows and columns in objective function
+ int nVars=0, nCons=0,temp1=0,temp2=0;
+ int numintcons=0;
+ double valobjsense;
+
+ CheckInputArgument(pvApiCtx , 11 , 11); //Checking the input arguments
+ CheckOutputArgument(pvApiCtx , 8, 8); //Checking the output arguments
+
+ ////////// Manage the input argument //////////
+
+ //Number of Variables
+ if(getIntFromScilab(1,&nVars))
+ {
+ return 1;
+ }
+
+ //Number of Constraints
+ if (getIntFromScilab(2,&nCons))
+ {
+ return 1;
+ }
+
+ //Objective function from Scilab
+ temp1 = nVars;
+ temp2 = nCons;
+ if (getFixedSizeDoubleMatrixFromScilab(3,1,temp1,&obj))
+ {
+ return 1;
+ }
+
+ //intcon matrix
+ if (getDoubleMatrixFromScilab(4,&numintcons,&temp2,&intcon))
+ {
+ return 1;
+ }
+
+ if (nCons!=0)
+ {
+ //conMatrix matrix from scilab
+ temp1 = nCons;
+ temp2 = nVars;
+
+ if (getFixedSizeDoubleMatrixFromScilab(5,temp1,temp2,&conMatrix))
+ {
+ return 1;
+ }
+
+ //conLB matrix from scilab
+ temp1 = nCons;
+ temp2 = 1;
+ if (getFixedSizeDoubleMatrixFromScilab(6,temp1,temp2,&conlb))
+ {
+ return 1;
+ }
+
+ //conUB matrix from scilab
+ if (getFixedSizeDoubleMatrixFromScilab(7,temp1,temp2,&conub))
+ {
+ return 1;
+ }
+
+ }
+
+ //lb matrix from scilab
+ temp1 = 1;
+ temp2 = nVars;
+ if (getFixedSizeDoubleMatrixFromScilab(8,temp1,temp2,&lb))
+ {
+ return 1;
+ }
+
+
+ //ub matrix from scilab
+ if (getFixedSizeDoubleMatrixFromScilab(9,temp1,temp2,&ub))
+ {
+ return 1;
+ }
+
+ //Object Sense
+ if(getDoubleFromScilab(10,&valobjsense))
+ {
+ return 1;
+ }
+
+ //get options from scilab
+ if(getFixedSizeDoubleMatrixFromScilab(11 , 1 , 4 , &options))
+ {
+ return 1;
+ }
+
+ //------------Temporary Version to make coin packed matrix------
+ OsiClpSolverInterface solver1;
+
+ CoinPackedMatrix *matrix = new CoinPackedMatrix(false , 0 , 0);
+ matrix->setDimensions(0 , nVars);
+ for(int i=0 ; i<nCons ; i++)
+ {
+ CoinPackedVector row;
+ for(int j=0 ; j<nVars ; j++)
+ {
+ row.insert(j, conMatrix[i+j*nCons]);
+ }
+ matrix->appendRow(row);
+ }
+
+
+ solver1.loadProblem(*matrix, lb, ub, obj, conlb, conub);
+
+ for(int i=0;i<numintcons;i++)
+ solver1.setInteger(intcon[i]-1);
+
+ solver1.setObjSense(valobjsense);
+
+ //-------------------------------------------------------------
+
+ CbcModel model(solver1);
+
+ model.solver()->setHintParam(OsiDoReducePrint, true, OsiHintTry);
+
+ if((int)options[0]!=0)
+ model.setIntegerTolerance(options[0]);
+ if((int)options[1]!=0)
+ model.setMaximumNodes((int)options[1]);
+ if((int)options[2]!=0)
+ model.setMaximumSeconds(options[2]);
+ if((int)options[3]!=0)
+ model.setAllowableGap(options[3]);
+
+ model.branchAndBound();
+
+ const double *val = model.getColSolution();
+
+ //Output the solution to Scilab
+
+ //get solution for x
+ double* xValue = model.getColSolution();
+
+ //get objective value
+ double objValue = model.getObjValue();
+
+ //Output status
+ double status_=-1;
+ if(model.isProvenOptimal()){
+ status_=0;
+ }
+ else if(model.isProvenInfeasible()){
+ status_=1;
+ }
+ else if(model.isSolutionLimitReached()){
+ status_=2;
+ }
+ else if(model. isNodeLimitReached()){
+ status_=3;
+ }
+ else if(model.isAbandoned()){
+ status_=4;
+ }
+ else if(model.isSecondsLimitReached()){
+ status_=5;
+ }
+ else if(model.isContinuousUnbounded()){
+ status_=6;
+ }
+ else if(model.isProvenDualInfeasible()){
+ status_=7;
+ }
+
+ double nodeCount=model.getNodeCount();
+ double nfps=model.numberIntegers();
+ double U=model.getObjValue();
+ double L=model.getBestPossibleObjValue();
+ double iterCount=model.getIterationCount();
+
+ returnDoubleMatrixToScilab(1 , nVars, 1 , xValue);
+ returnDoubleMatrixToScilab(2 , 1 , 1 , &objValue);
+ returnDoubleMatrixToScilab(3 , 1 , 1 , &status_);
+ returnDoubleMatrixToScilab(4 , 1 , 1 , &nodeCount);
+ returnDoubleMatrixToScilab(5 , 1 , 1 , &nfps);
+ returnDoubleMatrixToScilab(6 , 1 , 1 , &L);
+ returnDoubleMatrixToScilab(7 , 1 , 1 , &U);
+ returnDoubleMatrixToScilab(8 , 1 , 1 , &iterCount);
+
+ //-------------------------------------------------------------
+
+ return 0;
+}
+}
diff --git a/sci_gateway/cpp/cbcintlinprog_mpscpp.cpp b/sci_gateway/cpp/cbcintlinprog_mpscpp.cpp
new file mode 100644
index 0000000..8292ab1
--- /dev/null
+++ b/sci_gateway/cpp/cbcintlinprog_mpscpp.cpp
@@ -0,0 +1,115 @@
+// MILP with CBC library, mps
+// Finds the solution by using CBC Library
+// Code Authors: Akshay Miterani and Pranav Deshpande
+
+#include <sci_iofunc.hpp>
+
+// For Branch and bound
+#include "OsiSolverInterface.hpp"
+#include "CbcModel.hpp"=
+#include "CbcCutGenerator.hpp"
+#include "CbcHeuristicLocal.hpp"
+#include "OsiClpSolverInterface.hpp"
+extern "C" {
+#include <api_scilab.h>
+
+int mps_cppintlinprog()
+{
+ OsiClpSolverInterface solver;
+
+ // Path to the MPS file
+ char *mpsFilePath;
+
+ // Options to set maximum iterations
+ double *options;
+
+ // Input - 1 or 2 arguments allowed.
+ CheckInputArgument(pvApiCtx, 2, 2);
+
+ // Get the MPS File Path from Scilab
+ getStringFromScilab(1, &mpsFilePath);
+
+ // Receive the options for setting the maximum number of iterations etc.
+ if( getFixedSizeDoubleMatrixFromScilab(2, 1, 4, &options))
+ {
+ return 1;
+ }
+
+ // Read the MPS file
+ solver.readMps(mpsFilePath);
+
+ // Cbc Library used from here
+ CbcModel model(solver);
+
+ model.solver()->setHintParam(OsiDoReducePrint, true, OsiHintTry);
+
+ if((int)options[0]!=0)
+ model.setIntegerTolerance(options[0]);
+ if((int)options[1]!=0)
+ model.setMaximumNodes((int)options[1]);
+ if((int)options[2]!=0)
+ model.setMaximumSeconds(options[2]);
+ if((int)options[3]!=0)
+ model.setAllowableGap(options[3]);
+
+ model.branchAndBound();
+
+ int nVars = model.getNumCols();
+ int nCons = model.getNumRows();
+
+ const double *val = model.getColSolution();
+
+ //Output the solution to Scilab
+
+ //get solution for x
+ double* xValue = model.getColSolution();
+
+ //get objective value
+ double objValue = model.getObjValue();
+
+ //Output status
+ double status_=-1;
+ if(model.isProvenOptimal()){
+ status_=0;
+ }
+ else if(model.isProvenInfeasible()){
+ status_=1;
+ }
+ else if(model.isSolutionLimitReached()){
+ status_=2;
+ }
+ else if(model. isNodeLimitReached()){
+ status_=3;
+ }
+ else if(model.isAbandoned()){
+ status_=4;
+ }
+ else if(model.isSecondsLimitReached()){
+ status_=5;
+ }
+ else if(model.isContinuousUnbounded()){
+ status_=6;
+ }
+ else if(model.isProvenDualInfeasible()){
+ status_=7;
+ }
+
+ double nodeCount = model.getNodeCount();
+ double nfps = model.numberIntegers();
+ double U = model.getObjValue();
+ double L = model.getBestPossibleObjValue();
+ double iterCount = model.getIterationCount();
+
+ returnDoubleMatrixToScilab(1 , nVars, 1 , xValue);
+ returnDoubleMatrixToScilab(2 , 1 , 1 , &objValue);
+ returnDoubleMatrixToScilab(3 , 1 , 1 , &status_);
+ returnDoubleMatrixToScilab(4 , 1 , 1 , &nodeCount);
+ returnDoubleMatrixToScilab(5 , 1 , 1 , &nfps);
+ returnDoubleMatrixToScilab(6 , 1 , 1 , &L);
+ returnDoubleMatrixToScilab(7 , 1 , 1 , &U);
+ returnDoubleMatrixToScilab(8 , 1 , 1 , &iterCount);
+
+ return 0;
+}
+
+}
diff --git a/sci_gateway/cpp/cleaner.sce b/sci_gateway/cpp/cleaner.sce
new file mode 100755
index 0000000..333775c
--- /dev/null
+++ b/sci_gateway/cpp/cleaner.sce
@@ -0,0 +1,22 @@
+// This file is released under the 3-clause BSD license. See COPYING-BSD.
+// Generated by builder.sce : Please, do not edit this file
+// cleaner.sce
+// ------------------------------------------------------
+curdir = pwd();
+cleaner_path = get_file_path('cleaner.sce');
+chdir(cleaner_path);
+// ------------------------------------------------------
+if fileinfo('loader.sce') <> [] then
+ mdelete('loader.sce');
+end
+// ------------------------------------------------------
+if fileinfo('libFOSSEE_Optimization_Toolbox.so') <> [] then
+ mdelete('libFOSSEE_Optimization_Toolbox.so');
+end
+// ------------------------------------------------------
+if fileinfo('libFOSSEE_Optimization_Toolbox.c') <> [] then
+ mdelete('libFOSSEE_Optimization_Toolbox.c');
+end
+// ------------------------------------------------------
+chdir(curdir);
+// ------------------------------------------------------
diff --git a/sci_gateway/cpp/cpp_intfminbnd.cpp b/sci_gateway/cpp/cpp_intfminbnd.cpp
new file mode 100644
index 0000000..4914111
--- /dev/null
+++ b/sci_gateway/cpp/cpp_intfminbnd.cpp
@@ -0,0 +1,172 @@
+// Copyright (C) 2016 - IIT Bombay - FOSSEE
+//
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+// Author: Harpreet Singh
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+
+#include "CoinPragma.hpp"
+#include "CoinTime.hpp"
+#include "CoinError.hpp"
+
+#include "BonOsiTMINLPInterface.hpp"
+#include "BonIpoptSolver.hpp"
+#include "minbndTMINLP.hpp"
+#include "BonCbc.hpp"
+#include "BonBonminSetup.hpp"
+
+#include "BonOACutGenerator2.hpp"
+#include "BonEcpCuts.hpp"
+#include "BonOaNlpOptim.hpp"
+
+#include "sci_iofunc.hpp"
+extern "C"
+{
+#include "call_scilab.h"
+#include <api_scilab.h>
+#include <Scierror.h>
+#include <BOOL.h>
+#include <localization.h>
+#include <sciprint.h>
+
+int cpp_intfminbnd(char *fname)
+{
+ using namespace Ipopt;
+ using namespace Bonmin;
+
+ CheckInputArgument(pvApiCtx, 8, 8);
+ CheckOutputArgument(pvApiCtx, 3, 3);
+
+ // Input arguments
+ Number *integertolerance=NULL, *maxnodes=NULL, *allowablegap=NULL, *cputime=NULL,*max_iter=NULL, *lb = NULL, *ub = NULL;
+ static unsigned int nVars = 0;
+ unsigned int temp1 = 0,temp2 = 0, iret = 0;
+ int x0_rows, x0_cols,intconSize;
+ Number *intcon = NULL,*options=NULL, *ifval=NULL;
+
+ // Output arguments
+ Number *fX = NULL, ObjVal=0,iteration=0,cpuTime=0,fobj_eval=0;
+ Number dual_inf, constr_viol, complementarity, kkt_error;
+ int rstatus = 0;
+
+ if(getDoubleMatrixFromScilab(4, &x0_rows, &x0_cols, &lb))
+ {
+ return 1;
+ }
+
+ if(getDoubleMatrixFromScilab(5, &x0_rows, &x0_cols, &ub))
+ {
+ return 1;
+ }
+
+ // Getting intcon
+ if (getDoubleMatrixFromScilab(6,&intconSize,&temp2,&intcon))
+ {
+ return 1;
+ }
+
+ //Initialization of parameters
+ nVars=x0_rows;
+ temp1 = 1;
+ temp2 = 1;
+
+ //Getting parameters
+ if (getFixedSizeDoubleMatrixInList(7,2,temp1,temp2,&integertolerance))
+ {
+ return 1;
+ }
+ if (getFixedSizeDoubleMatrixInList(7,4,temp1,temp2,&maxnodes))
+ {
+ return 1;
+ }
+ if (getFixedSizeDoubleMatrixInList(7,6,temp1,temp2,&cputime))
+ {
+ return 1;
+ }
+ if (getFixedSizeDoubleMatrixInList(7,8,temp1,temp2,&allowablegap))
+ {
+ return 1;
+ }
+ if (getFixedSizeDoubleMatrixInList(7,10,temp1,temp2,&max_iter))
+ {
+ return 1;
+ }
+
+ SmartPtr<minbndTMINLP> tminlp = new minbndTMINLP(nVars,lb,ub,intconSize,intcon);
+
+ BonminSetup bonmin;
+ bonmin.initializeOptionsAndJournalist();
+
+ bonmin.options()->SetStringValue("mu_oracle","loqo");
+ bonmin.options()->SetNumericValue("bonmin.integer_tolerance", *integertolerance);
+ bonmin.options()->SetIntegerValue("bonmin.node_limit", (int)*maxnodes);
+ bonmin.options()->SetNumericValue("bonmin.time_limit", *cputime);
+ bonmin.options()->SetNumericValue("bonmin.allowable_gap", *allowablegap);
+ bonmin.options()->SetIntegerValue("bonmin.iteration_limit", (int)*max_iter);
+
+ //Now initialize from tminlp
+ bonmin.initialize(GetRawPtr(tminlp));
+
+ //Set up done, now let's branch and bound
+ try {
+ Bab bb;
+ bb(bonmin);//process parameter file using Ipopt and do branch and bound using Cbc
+ }
+ catch(TNLPSolver::UnsolvedError *E) {
+ Scierror(999, "\nIpopt has failed to solve the problem!\n");
+ }
+ catch(OsiTMINLPInterface::SimpleError &E) {
+ Scierror(999, "\nFailed to solve a problem!\n");
+ }
+ catch(CoinError &E) {
+ Scierror(999, "\nFailed to solve a problem!\n");
+ }
+ rstatus=tminlp->returnStatus();
+
+ if(rstatus==0 ||rstatus== 3)
+ {
+ fX = tminlp->getX();
+ ObjVal = tminlp->getObjVal();
+ if (returnDoubleMatrixToScilab(1, nVars, 1, fX))
+ {
+ return 1;
+ }
+
+ if (returnDoubleMatrixToScilab(2, 1, 1, &ObjVal))
+ {
+ return 1;
+ }
+
+ if (returnIntegerMatrixToScilab(3, 1, 1, &rstatus))
+ {
+ return 1;
+ }
+
+ }
+ else
+ {
+ if (returnDoubleMatrixToScilab(1, 0, 0, fX))
+ {
+ return 1;
+ }
+
+ if (returnDoubleMatrixToScilab(2, 1, 1, &ObjVal))
+ {
+ return 1;
+ }
+
+ if (returnIntegerMatrixToScilab(3, 1, 1, &rstatus))
+ {
+ return 1;
+ }
+
+ }
+
+ return 0;
+ }
+}
+
diff --git a/sci_gateway/cpp/cpp_intfmincon.cpp b/sci_gateway/cpp/cpp_intfmincon.cpp
new file mode 100644
index 0000000..d921128
--- /dev/null
+++ b/sci_gateway/cpp/cpp_intfmincon.cpp
@@ -0,0 +1,189 @@
+// Copyright (C) 2016 - IIT Bombay - FOSSEE
+//
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+// Author: Harpreet Singh
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+
+#include "CoinPragma.hpp"
+#include "CoinTime.hpp"
+#include "CoinError.hpp"
+
+#include "BonOsiTMINLPInterface.hpp"
+#include "BonIpoptSolver.hpp"
+#include "minconTMINLP.hpp"
+#include "BonCbc.hpp"
+#include "BonBonminSetup.hpp"
+
+#include "BonOACutGenerator2.hpp"
+#include "BonEcpCuts.hpp"
+#include "BonOaNlpOptim.hpp"
+
+#include "sci_iofunc.hpp"
+extern "C"
+{
+#include "call_scilab.h"
+#include <api_scilab.h>
+#include <Scierror.h>
+#include <BOOL.h>
+#include <localization.h>
+#include <sciprint.h>
+
+int cpp_intfmincon(char *fname)
+{
+ using namespace Ipopt;
+ using namespace Bonmin;
+
+ CheckInputArgument(pvApiCtx, 13, 13);
+ CheckOutputArgument(pvApiCtx, 3, 3);
+
+ // Input arguments
+ Number *integertolerance=NULL, *maxnodes=NULL, *allowablegap=NULL, *cputime=NULL,*max_iter=NULL;
+ Number *x0 = NULL, *lb = NULL, *ub = NULL,*conLb = NULL, *conUb = NULL,*LC = NULL;
+ static unsigned int nVars = 0,nCons = 0;
+ unsigned int temp1 = 0,temp2 = 0, iret = 0;
+ int x0_rows, x0_cols,intconSize;
+ Number *intcon = NULL,*options=NULL, *ifval=NULL;
+
+ // Output arguments
+ Number *fX = NULL, ObjVal=0,iteration=0,cpuTime=0,fobj_eval=0;
+ Number dual_inf, constr_viol, complementarity, kkt_error;
+ int rstatus = 0;
+
+ if(getDoubleMatrixFromScilab(6, &nVars, &x0_cols, &x0))
+ {
+ return 1;
+ }
+
+ if(getDoubleMatrixFromScilab(7, &x0_rows, &x0_cols, &lb))
+ {
+ return 1;
+ }
+
+ if(getDoubleMatrixFromScilab(8, &x0_rows, &x0_cols, &ub))
+ {
+ return 1;
+ }
+
+ if(getDoubleMatrixFromScilab(9, &nCons, &x0_cols, &conLb))
+ {
+ return 1;
+ }
+
+ if(getDoubleMatrixFromScilab(10, &x0_rows, &x0_cols, &conUb))
+ {
+ return 1;
+ }
+
+ // Getting intcon
+ if (getDoubleMatrixFromScilab(11,&intconSize,&temp2,&intcon))
+ {
+ return 1;
+ }
+
+ if (getDoubleMatrixFromScilab(13,&temp1,&temp2,&LC))
+ {
+ return 1;
+ }
+
+ //Initialization of parameters
+ temp1 = 1;
+ temp2 = 1;
+
+ //Getting parameters
+ if (getFixedSizeDoubleMatrixInList(12,2,temp1,temp2,&integertolerance))
+ {
+ return 1;
+ }
+ if (getFixedSizeDoubleMatrixInList(12,4,temp1,temp2,&maxnodes))
+ {
+ return 1;
+ }
+ if (getFixedSizeDoubleMatrixInList(12,6,temp1,temp2,&cputime))
+ {
+ return 1;
+ }
+ if (getFixedSizeDoubleMatrixInList(12,8,temp1,temp2,&allowablegap))
+ {
+ return 1;
+ }
+ if (getFixedSizeDoubleMatrixInList(12,10,temp1,temp2,&max_iter))
+ {
+ return 1;
+ }
+
+ SmartPtr<minconTMINLP> tminlp = new minconTMINLP(nVars,x0,lb,ub,(unsigned int)LC,nCons,conLb,conUb,intconSize,intcon);
+
+ BonminSetup bonmin;
+ bonmin.initializeOptionsAndJournalist();
+ bonmin.options()->SetStringValue("mu_oracle","loqo");
+ bonmin.options()->SetIntegerValue("bonmin.print_level",5);
+ bonmin.options()->SetNumericValue("bonmin.integer_tolerance", *integertolerance);
+ bonmin.options()->SetIntegerValue("bonmin.node_limit", (int)*maxnodes);
+ bonmin.options()->SetNumericValue("bonmin.time_limit", *cputime);
+ bonmin.options()->SetNumericValue("bonmin.allowable_gap", *allowablegap);
+ bonmin.options()->SetIntegerValue("bonmin.iteration_limit", (int)*max_iter);
+
+ //Now initialize from tminlp
+ bonmin.initialize(GetRawPtr(tminlp));
+
+ //Set up done, now let's branch and bound
+ try {
+ Bab bb;
+ bb(bonmin);//process parameter file using Ipopt and do branch and bound using Cbc
+ }
+ catch(TNLPSolver::UnsolvedError *E) {
+ }
+ catch(OsiTMINLPInterface::SimpleError &E) {
+ }
+ catch(CoinError &E) {
+ }
+ rstatus=tminlp->returnStatus();
+
+ if(rstatus==0 ||rstatus== 3)
+ {
+ fX = tminlp->getX();
+ ObjVal = tminlp->getObjVal();
+ if (returnDoubleMatrixToScilab(1, nVars, 1, fX))
+ {
+ return 1;
+ }
+
+ if (returnDoubleMatrixToScilab(2, 1, 1, &ObjVal))
+ {
+ return 1;
+ }
+
+ if (returnIntegerMatrixToScilab(3, 1, 1, &rstatus))
+ {
+ return 1;
+ }
+
+ }
+ else
+ {
+ if (returnDoubleMatrixToScilab(1, 0, 0, fX))
+ {
+ return 1;
+ }
+
+ if (returnDoubleMatrixToScilab(2, 1, 1, &ObjVal))
+ {
+ return 1;
+ }
+
+ if (returnIntegerMatrixToScilab(3, 1, 1, &rstatus))
+ {
+ return 1;
+ }
+
+ }
+
+ return 0;
+ }
+}
+
diff --git a/sci_gateway/cpp/cpp_intfminunc.cpp b/sci_gateway/cpp/cpp_intfminunc.cpp
new file mode 100644
index 0000000..3e1abcd
--- /dev/null
+++ b/sci_gateway/cpp/cpp_intfminunc.cpp
@@ -0,0 +1,174 @@
+// Copyright (C) 2016 - IIT Bombay - FOSSEE
+//
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+// Author: Harpreet Singh, Pranav Deshpande and Akshay Miterani
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+
+#include "CoinPragma.hpp"
+#include "CoinTime.hpp"
+#include "CoinError.hpp"
+
+#include "BonOsiTMINLPInterface.hpp"
+#include "BonIpoptSolver.hpp"
+#include "minuncTMINLP.hpp"
+#include "BonCbc.hpp"
+#include "BonBonminSetup.hpp"
+
+#include "BonOACutGenerator2.hpp"
+#include "BonEcpCuts.hpp"
+#include "BonOaNlpOptim.hpp"
+
+#include "sci_iofunc.hpp"
+extern "C"
+{
+#include "call_scilab.h"
+#include <api_scilab.h>
+#include <Scierror.h>
+#include <BOOL.h>
+#include <localization.h>
+#include <sciprint.h>
+
+int cpp_intfminunc(char *fname)
+{
+ using namespace Ipopt;
+ using namespace Bonmin;
+
+ CheckInputArgument(pvApiCtx, 8, 8);
+ CheckOutputArgument(pvApiCtx, 3, 3); // 3 output arguments
+
+ //Function pointers, input matrix(Starting point) pointer, flag variable
+ int* funptr=NULL;
+ double* x0ptr=NULL;
+
+ // Input arguments
+ Number *integertolerance=NULL, *maxnodes=NULL, *allowablegap=NULL, *cputime=NULL,*max_iter=NULL;
+ static unsigned int nVars = 0,nCons = 0;
+ unsigned int temp1 = 0,temp2 = 0, iret = 0;
+ int x0_rows, x0_cols;
+ double *intcon = NULL,*options=NULL, *ifval=NULL;
+ int intconSize;
+
+ // Output arguments
+ double *fX = NULL, ObjVal=0,iteration=0,cpuTime=0,fobj_eval=0;
+ double dual_inf, constr_viol, complementarity, kkt_error;
+ int rstatus = 0;
+ int int_fobj_eval, int_constr_eval, int_fobj_grad_eval, int_constr_jac_eval, int_hess_eval;
+
+ //x0(starting point) matrix from scilab
+ if(getDoubleMatrixFromScilab(4, &x0_rows, &x0_cols, &x0ptr))
+ {
+ return 1;
+ }
+
+ nVars=x0_rows;
+
+ // Getting intcon
+ if (getDoubleMatrixFromScilab(5,&intconSize,&temp2,&intcon))
+ {
+ return 1;
+ }
+
+ temp1 = 1;
+ temp2 = 1;
+
+ //Getting parameters
+ if (getFixedSizeDoubleMatrixInList(6,2,temp1,temp2,&integertolerance))
+ {
+ return 1;
+ }
+ if (getFixedSizeDoubleMatrixInList(6,4,temp1,temp2,&maxnodes))
+ {
+ return 1;
+ }
+ if (getFixedSizeDoubleMatrixInList(6,6,temp1,temp2,&cputime))
+ {
+ return 1;
+ }
+ if (getFixedSizeDoubleMatrixInList(6,8,temp1,temp2,&allowablegap))
+ {
+ return 1;
+ }
+ if (getFixedSizeDoubleMatrixInList(6,10,temp1,temp2,&max_iter))
+ {
+ return 1;
+ }
+
+ SmartPtr<minuncTMINLP> tminlp = new minuncTMINLP(nVars, x0ptr, intconSize, intcon);
+
+ BonminSetup bonmin;
+ bonmin.initializeOptionsAndJournalist();
+
+ // Here we can change the default value of some Bonmin or Ipopt option
+ bonmin.options()->SetStringValue("mu_oracle","loqo");
+ bonmin.options()->SetNumericValue("bonmin.integer_tolerance", *integertolerance);
+ bonmin.options()->SetIntegerValue("bonmin.node_limit", (int)*maxnodes);
+ bonmin.options()->SetNumericValue("bonmin.time_limit", *cputime);
+ bonmin.options()->SetNumericValue("bonmin.allowable_gap", *allowablegap);
+ bonmin.options()->SetIntegerValue("bonmin.iteration_limit", (int)*max_iter);
+
+ //Now initialize from tminlp
+ bonmin.initialize(GetRawPtr(tminlp));
+
+ //Set up done, now let's branch and bound
+ try {
+ Bab bb;
+ bb(bonmin);//process parameter file using Ipopt and do branch and bound using Cbc
+ }
+ catch(TNLPSolver::UnsolvedError *E) {
+ //There has been a failure to solve a problem with Ipopt.
+ Scierror(999, "\nIpopt has failed to solve the problem!\n");
+ }
+ catch(OsiTMINLPInterface::SimpleError &E) {
+ Scierror(999, "\nFailed to solve a problem!\n");
+ }
+ catch(CoinError &E) {
+ Scierror(999, "\nFailed to solve a problem!\n");
+ }
+ rstatus=tminlp->returnStatus();
+ if(rstatus==0 ||rstatus== 3)
+ {
+ fX = tminlp->getX();
+ ObjVal = tminlp->getObjVal();
+ if (returnDoubleMatrixToScilab(1, nVars, 1, fX))
+ {
+ return 1;
+ }
+
+ if (returnDoubleMatrixToScilab(2, 1, 1, &ObjVal))
+ {
+ return 1;
+ }
+
+ if (returnIntegerMatrixToScilab(3, 1, 1, &rstatus))
+ {
+ return 1;
+ }
+
+ }
+ else
+ {
+ if (returnDoubleMatrixToScilab(1, 0, 0, fX))
+ {
+ return 1;
+ }
+
+ if (returnDoubleMatrixToScilab(2, 1, 1, &ObjVal))
+ {
+ return 1;
+ }
+
+ if (returnIntegerMatrixToScilab(3, 1, 1, &rstatus))
+ {
+ return 1;
+ }
+ }
+
+ return 0;
+ }
+}
+
diff --git a/sci_gateway/cpp/cpp_intqpipopt.cpp b/sci_gateway/cpp/cpp_intqpipopt.cpp
new file mode 100644
index 0000000..d89d643
--- /dev/null
+++ b/sci_gateway/cpp/cpp_intqpipopt.cpp
@@ -0,0 +1,267 @@
+// Copyright (C) 2016 - IIT Bombay - FOSSEE
+//
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+// Author: Harpreet Singh, Pranav Deshpande and Akshay Miterani
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+
+#include <iomanip>
+#include <fstream>
+#include <iostream>
+#include "CoinPragma.hpp"
+#include "CoinTime.hpp"
+#include "CoinError.hpp"
+
+#include "BonOsiTMINLPInterface.hpp"
+#include "BonIpoptSolver.hpp"
+#include "QuadTMINLP.hpp"
+#include "BonCbc.hpp"
+#include "BonBonminSetup.hpp"
+
+#include "BonOACutGenerator2.hpp"
+#include "BonEcpCuts.hpp"
+#include "BonOaNlpOptim.hpp"
+
+#include "sci_iofunc.hpp"
+extern "C"
+{
+#include <api_scilab.h>
+#include <Scierror.h>
+#include <BOOL.h>
+#include <localization.h>
+#include <sciprint.h>
+
+int cpp_intqpipopt(char *fname)
+{
+ using namespace Ipopt;
+ using namespace Bonmin;
+
+ CheckInputArgument(pvApiCtx, 15, 15); // We need total 15 input arguments.
+ CheckOutputArgument(pvApiCtx, 3, 3); // 3 output arguments
+
+ // Error management variable
+ SciErr sciErr;
+
+ // Input arguments
+ double *QItems=NULL,*PItems=NULL, *intcon = NULL, *ConItems=NULL,*conUB=NULL,*conLB=NULL;
+ double *varUB=NULL,*varLB=NULL,*init_guess=NULL,*options=NULL, *ifval=NULL;
+ static unsigned int nVars = 0,nCons = 0, intconSize = 0;
+ unsigned int temp1 = 0,temp2 = 0;
+ char *bonmin_options_file = NULL;
+ // Output arguments
+ double *fX = NULL, ObjVal = 0,iteration=0;
+ int rstatus = 0;
+
+ //Number of Variables
+ if(getIntFromScilab(1,&nVars))
+ {
+ return 1;
+ }
+
+ //Number of Constraints
+ if (getIntFromScilab(2,&nCons))
+ {
+ return 1;
+ }
+
+ //Number of variables constrained to be integers
+ if (getIntFromScilab(3,&intconSize))
+ {
+ return 1;
+ }
+
+ //Q matrix from scilab
+ temp1 = nVars;
+ temp2 = nVars;
+ if (getFixedSizeDoubleMatrixFromScilab(4,temp1,temp1,&QItems))
+ {
+ return 1;
+ }
+
+ //P matrix from scilab
+ temp1 = 1;
+ temp2 = nVars;
+ if (getFixedSizeDoubleMatrixFromScilab(5,temp1,temp2,&PItems))
+ {
+ return 1;
+ }
+
+ temp1 = 1;
+ temp2 = intconSize;
+ // Getting intcon
+ if (getDoubleMatrixFromScilab(6,&temp1,&temp2,&intcon))
+ {
+ return 1;
+ }
+
+ if (nCons!=0)
+ {
+ //conMatrix matrix from scilab
+ temp1 = nCons;
+ temp2 = nVars;
+
+ if (getFixedSizeDoubleMatrixFromScilab(7,temp1,temp2,&ConItems))
+ {
+ return 1;
+ }
+
+ //conLB matrix from scilab
+ temp1 = 1;
+ temp2 = nCons;
+ if (getFixedSizeDoubleMatrixFromScilab(8,temp1,temp2,&conLB))
+ {
+ return 1;
+ }
+
+ //conUB matrix from scilab
+ if (getFixedSizeDoubleMatrixFromScilab(9,temp1,temp2,&conUB))
+ {
+ return 1;
+ }
+ }
+
+ //varLB matrix from scilab
+ temp1 = 1;
+ temp2 = nVars;
+ if (getFixedSizeDoubleMatrixFromScilab(10,temp1,temp2,&varLB))
+ {
+ return 1;
+ }
+
+ //varUB matrix from scilab
+ if (getFixedSizeDoubleMatrixFromScilab(11,temp1,temp2,&varUB))
+ {
+ return 1;
+ }
+
+ //Initial Value of variables from scilab
+ if (getFixedSizeDoubleMatrixFromScilab( 12,temp1,temp2,&init_guess))
+ {
+ return 1;
+ }
+
+ temp1=1;
+ temp2=5;
+ if (getFixedSizeDoubleMatrixFromScilab(13,temp1,temp2,&options))
+ {
+ return 1;
+ }
+
+ temp1=1;
+ temp2=5;
+ if (getFixedSizeDoubleMatrixFromScilab(14,temp1,temp2,&ifval))
+ {
+ return 1;
+ }
+
+ if (getStringFromScilab(15, &bonmin_options_file))
+ {
+ return 1;
+ }
+
+
+
+
+ SmartPtr<QuadTMINLP> tminlp = new QuadTMINLP(nVars,nCons,intconSize,QItems, PItems, intcon,ConItems,conLB,conUB,varLB,varUB,init_guess);
+
+ BonminSetup bonmin;
+ bonmin.initializeOptionsAndJournalist();
+
+ // Here we can change the default value of some Bonmin or Ipopt option
+ bonmin.options()->SetStringValue("mu_oracle","loqo");
+
+
+ //Register an additional option
+ if((int)ifval[0])
+ bonmin.options()->SetNumericValue("bonmin.integer_tolerance", (options[0]));
+ if((int)ifval[1])
+ bonmin.options()->SetIntegerValue("bonmin.node_limit", (int)(options[1]));
+ if((int)ifval[2])
+ bonmin.options()->SetNumericValue("bonmin.time_limit", (options[2]));
+ if((int)ifval[3])
+ bonmin.options()->SetNumericValue("bonmin.allowable_gap", (options[3]));
+ if((int)ifval[4])
+ bonmin.options()->SetIntegerValue("bonmin.iteration_limit", (int)(options[4]));
+
+
+ //Here we read the option file
+ //if ( bonmin_options_file!=NULL )
+ // bonmin.readOptionsFile(bonmin_options_file);
+
+ //Now initialize from tminlp
+ bonmin.initialize(GetRawPtr(tminlp));
+
+ //Set up done, now let's branch and bound
+ try {
+ Bab bb;
+ bb(bonmin);//process parameter file using Ipopt and do branch and bound using Cbc
+ }
+ catch(TNLPSolver::UnsolvedError *E) {
+ //There has been a failure to solve a problem with Ipopt.
+ std::cerr<<"Ipopt has failed to solve a problem!"<<std::endl;
+ sciprint(999, "\nIpopt has failed to solve the problem!\n");
+ }
+ catch(OsiTMINLPInterface::SimpleError &E) {
+ std::cerr<<E.className()<<"::"<<E.methodName()
+ <<std::endl
+ <<E.message()<<std::endl;
+ sciprint(999, "\nFailed to solve a problem!\n");
+ }
+ catch(CoinError &E) {
+ std::cerr<<E.className()<<"::"<<E.methodName()
+ <<std::endl
+ <<E.message()<<std::endl;
+ sciprint(999, "\nFailed to solve a problem!\n");
+ }
+ rstatus=tminlp->returnStatus();
+ if (rstatus >= 0 | rstatus <= 5){
+ fX = tminlp->getX();
+ ObjVal = tminlp->getObjVal();
+ if (returnDoubleMatrixToScilab(1, 1, nVars, fX))
+ {
+ return 1;
+ }
+
+ if (returnDoubleMatrixToScilab(2, 1, 1, &ObjVal))
+ {
+ return 1;
+ }
+
+ if (returnIntegerMatrixToScilab(3, 1, 1, &rstatus))
+ {
+ return 1;
+ }
+
+ }
+ else
+ {
+ if (returnDoubleMatrixToScilab(1, 0, 0, fX))
+ {
+ return 1;
+ }
+
+ if (returnDoubleMatrixToScilab(2, 1, 1, &ObjVal))
+ {
+ return 1;
+ }
+
+ if (returnIntegerMatrixToScilab(3, 1, 1, &rstatus))
+ {
+ return 1;
+ }
+
+ sciprint(999, "\nThe problem could not be solved!\n");
+ }
+
+ // As the SmartPtrs go out of scope, the reference count
+ // will be decremented and the objects will automatically
+ // be deleted(No memory leakage).
+
+ return 0;
+}
+}
+
diff --git a/sci_gateway/cpp/libFOSSEE_Optimization_Toolbox.c b/sci_gateway/cpp/libFOSSEE_Optimization_Toolbox.c
new file mode 100644
index 0000000..7090628
--- /dev/null
+++ b/sci_gateway/cpp/libFOSSEE_Optimization_Toolbox.c
@@ -0,0 +1,40 @@
+#ifdef __cplusplus
+extern "C" {
+#endif
+#include <mex.h>
+#include <sci_gateway.h>
+#include <api_scilab.h>
+#include <MALLOC.h>
+static int direct_gateway(char *fname,void F(void)) { F();return 0;};
+extern Gatefunc cpp_intfminunc;
+extern Gatefunc cpp_intfminbnd;
+extern Gatefunc cpp_intfmincon;
+extern Gatefunc cpp_intqpipopt;
+extern Gatefunc matrix_cppintlinprog;
+extern Gatefunc mps_cppintlinprog;
+static GenericTable Tab[]={
+ {(Myinterfun)sci_gateway,cpp_intfminunc,"inter_fminunc"},
+ {(Myinterfun)sci_gateway,cpp_intfminbnd,"inter_fminbnd"},
+ {(Myinterfun)sci_gateway,cpp_intfmincon,"inter_fmincon"},
+ {(Myinterfun)sci_gateway,cpp_intqpipopt,"sci_intqpipopt"},
+ {(Myinterfun)sci_gateway,matrix_cppintlinprog,"sci_matrix_intlinprog"},
+ {(Myinterfun)sci_gateway,mps_cppintlinprog,"sci_mps_intlinprog"},
+};
+
+int C2F(libFOSSEE_Optimization_Toolbox)()
+{
+ Rhs = Max(0, Rhs);
+ if (*(Tab[Fin-1].f) != NULL)
+ {
+ if(pvApiCtx == NULL)
+ {
+ pvApiCtx = (StrCtx*)MALLOC(sizeof(StrCtx));
+ }
+ pvApiCtx->pstName = (char*)Tab[Fin-1].name;
+ (*(Tab[Fin-1].f))(Tab[Fin-1].name,Tab[Fin-1].F);
+ }
+ return 0;
+}
+#ifdef __cplusplus
+}
+#endif
diff --git a/sci_gateway/cpp/libFOSSEE_Optimization_Toolbox.so b/sci_gateway/cpp/libFOSSEE_Optimization_Toolbox.so
new file mode 100755
index 0000000..233098e
--- /dev/null
+++ b/sci_gateway/cpp/libFOSSEE_Optimization_Toolbox.so
Binary files differ
diff --git a/sci_gateway/cpp/loader.sce b/sci_gateway/cpp/loader.sce
new file mode 100644
index 0000000..cad0490
--- /dev/null
+++ b/sci_gateway/cpp/loader.sce
@@ -0,0 +1,26 @@
+// This file is released under the 3-clause BSD license. See COPYING-BSD.
+// Generated by builder.sce : Please, do not edit this file
+// ----------------------------------------------------------------------------
+//
+libFOSSEE_Optimizat_path = get_absolute_file_path('loader.sce');
+//
+// ulink previous function with same name
+[bOK, ilib] = c_link('libFOSSEE_Optimization_Toolbox');
+if bOK then
+ ulink(ilib);
+end
+//
+list_functions = [ 'inter_fminunc';
+ 'inter_fminbnd';
+ 'inter_fmincon';
+ 'sci_intqpipopt';
+ 'sci_matrix_intlinprog';
+ 'sci_mps_intlinprog';
+];
+addinter(libFOSSEE_Optimizat_path + filesep() + 'libFOSSEE_Optimization_Toolbox' + getdynlibext(), 'libFOSSEE_Optimization_Toolbox', list_functions);
+// remove temp. variables on stack
+clear libFOSSEE_Optimizat_path;
+clear bOK;
+clear ilib;
+clear list_functions;
+// ----------------------------------------------------------------------------
diff --git a/sci_gateway/cpp/minbndTMINLP.hpp b/sci_gateway/cpp/minbndTMINLP.hpp
new file mode 100644
index 0000000..581d5ce
--- /dev/null
+++ b/sci_gateway/cpp/minbndTMINLP.hpp
@@ -0,0 +1,114 @@
+// Copyright (C) 2016 - IIT Bombay - FOSSEE
+//
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+// Author: Harpreet Singh
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+
+#ifndef minbndTMINLP_HPP
+#define minbndTMINLP_HPP
+
+#include "BonTMINLP.hpp"
+#include "IpTNLP.hpp"
+#include "call_scilab.h"
+
+using namespace Ipopt;
+using namespace Bonmin;
+
+class minbndTMINLP : public TMINLP
+{
+ private:
+
+ Index numVars_; //Number of input variables
+
+ Index intconSize_;
+
+ Number *lb_= NULL; //lb_ is a pointer to a matrix of size of 1*numVars_ with lower bound of all variables.
+
+ Number *ub_= NULL; //ub_ is a pointer to a matrix of size of 1*numVars_ with upper bound of all variables.
+
+ Number *finalX_= NULL; //finalX_ is a pointer to a matrix of size of 1*numVars_ with final value for the primal variables.
+
+ Number finalObjVal_; //finalObjVal_ is a scalar with the final value of the objective.
+
+ Number *intcon_ = NULL;
+
+ int status_; //Solver return status
+ minbndTMINLP(const minbndTMINLP&);
+ minbndTMINLP& operator=(const minbndTMINLP&);
+
+public:
+ // Constructor
+ minbndTMINLP(Index nV, Number *lb, Number *ub, Index intconSize, Number *intcon):numVars_(nV),lb_(lb),ub_(ub),intconSize_(intconSize),intcon_(intcon),finalX_(0),finalObjVal_(1e20){ }
+
+ /** default destructor */
+ virtual ~minbndTMINLP();
+
+ virtual bool get_variables_types(Index n, VariableType* var_types);
+
+ virtual bool get_variables_linearity(Index n, Ipopt::TNLP::LinearityType* var_types);
+
+ virtual bool get_constraints_linearity(Index m, Ipopt::TNLP::LinearityType* const_types);
+
+ /** Method to return some info about the nlp */
+ virtual bool get_nlp_info(Index& n, Index& m, Index& nnz_jac_g,
+ Index& nnz_h_lag, TNLP::IndexStyleEnum& index_style);
+
+ /** Method to return the bounds for my problem */
+ virtual bool get_bounds_info(Index n, Number* x_l, Number* x_u,
+ Index m, Number* g_l, Number* g_u);
+
+ /** Method to return the starting point for the algorithm */
+ virtual bool get_starting_point(Index n, bool init_x, Number* x,
+ bool init_z, Number* z_L, Number* z_U,
+ Index m, bool init_lambda,
+ Number* lambda);
+
+ /** Method to return the objective value */
+ virtual bool eval_f(Index n, const Number* x, bool new_x, Number& obj_value);
+
+ /** Method to return the gradient of the objective */
+ virtual bool eval_grad_f(Index n, const Number* x, bool new_x, Number* grad_f);
+
+ /** Method to return the constraint residuals */
+ virtual bool eval_g(Index n, const Number* x, bool new_x, Index m, Number* g);
+
+ /** Method to return:
+ * 1) The structure of the jacobian (if "values" is NULL)
+ * 2) The values of the jacobian (if "values" is not NULL)
+ */
+ virtual bool eval_jac_g(Index n, const Number* x, bool new_x,Index m, Index nele_jac, Index* iRow, Index *jCol,Number* values);
+
+ /** Method to return:
+ * 1) The structure of the hessian of the lagrangian (if "values" is NULL)
+ * 2) The values of the hessian of the lagrangian (if "values" is not NULL)
+ */
+ virtual bool eval_h(Index n, const Number* x, bool new_x,Number obj_factor, Index m, const Number* lambda,bool new_lambda, Index nele_hess, Index* iRow,Index* jCol, Number* values);
+
+ /** This method is called when the algorithm is complete so the TNLP can store/write the solution */
+ virtual void finalize_solution(SolverReturn status,Index n, const Number* x, Number obj_value);
+
+ virtual const SosInfo * sosConstraints() const{return NULL;}
+ virtual const BranchingInfo* branchingInfo() const{return NULL;}
+
+ const double * getX(); //Returns a pointer to a matrix of size of 1*numVars_
+ //with final value for the primal variables.
+
+ const double * getGrad(); //Returns a pointer to a matrix of size of 1*numVars_
+ //with final value of gradient for the primal variables.
+
+ const double * getHess(); //Returns a pointer to a matrix of size of numVars_*numVars_
+ //with final value of hessian for the primal variables.
+
+ double getObjVal(); //Returns the output of the final value of the objective.
+
+ double iterCount(); //Returns the iteration count
+
+ int returnStatus(); //Returns the status count
+};
+
+#endif
diff --git a/sci_gateway/cpp/minconTMINLP.hpp b/sci_gateway/cpp/minconTMINLP.hpp
new file mode 100644
index 0000000..5b3006a
--- /dev/null
+++ b/sci_gateway/cpp/minconTMINLP.hpp
@@ -0,0 +1,124 @@
+// Copyright (C) 2016 - IIT Bombay - FOSSEE
+//
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+// Author: Harpreet Singh, Pranav Deshpande and Akshay Miterani
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+
+#ifndef minconTMINLP_HPP
+#define minconTMINLP_HPP
+
+#include "BonTMINLP.hpp"
+#include "IpTNLP.hpp"
+#include "call_scilab.h"
+
+using namespace Ipopt;
+using namespace Bonmin;
+
+class minconTMINLP : public TMINLP
+{
+ private:
+
+ Index numVars_; //Number of variables
+
+ Index numCons_; //Number of constraints
+
+ Index numLC_; //Number of Linear constraints
+
+ Index intconSize_;
+
+ Number *x0_= NULL; //lb_ is a pointer to a matrix of size of 1*numVars_ with lower bound of all variables.
+
+ Number *lb_= NULL; //lb_ is a pointer to a matrix of size of 1*numVars_ with lower bound of all variables.
+
+ Number *ub_= NULL; //ub_ is a pointer to a matrix of size of 1*numVars_ with upper bound of all variables.
+
+ Number *conLb_= NULL; //conLb_ is a pointer to a matrix of size of numCon_*1 with lower bound of all constraints.
+
+ Number *conUb_= NULL; //conUb_ is a pointer to a matrix of size of numCon_*1 with upper bound of all constraints.
+
+ Number *finalX_= NULL; //finalX_ is a pointer to a matrix of size of 1*numVars_ with final value for the primal variables.
+
+ Number finalObjVal_; //finalObjVal_ is a scalar with the final value of the objective.
+
+ Number *intcon_ = NULL;
+
+ int status_; //Solver return status
+ minconTMINLP(const minconTMINLP&);
+ minconTMINLP& operator=(const minconTMINLP&);
+
+public:
+ // Constructor
+ minconTMINLP(Index nV, Number *x0, Number *lb, Number *ub, Index nLC, Index nCons, Number *conlb, Number *conub, Index intconSize, Number *intcon):numVars_(nV),x0_(x0),lb_(lb),ub_(ub),numLC_(nLC),numCons_(nCons),conLb_(conlb),conUb_(conub),intconSize_(intconSize),intcon_(intcon),finalX_(0),finalObjVal_(1e20){ }
+
+ /** default destructor */
+ virtual ~minconTMINLP();
+
+ virtual bool get_variables_types(Index n, VariableType* var_types);
+
+ virtual bool get_variables_linearity(Index n, Ipopt::TNLP::LinearityType* var_types);
+
+ virtual bool get_constraints_linearity(Index m, Ipopt::TNLP::LinearityType* const_types);
+
+ /** Method to return some info about the nlp */
+ virtual bool get_nlp_info(Index& n, Index& m, Index& nnz_jac_g,
+ Index& nnz_h_lag, TNLP::IndexStyleEnum& index_style);
+
+ /** Method to return the bounds for my problem */
+ virtual bool get_bounds_info(Index n, Number* x_l, Number* x_u,
+ Index m, Number* g_l, Number* g_u);
+
+ /** Method to return the starting point for the algorithm */
+ virtual bool get_starting_point(Index n, bool init_x, Number* x,
+ bool init_z, Number* z_L, Number* z_U,
+ Index m, bool init_lambda,
+ Number* lambda);
+
+ /** Method to return the objective value */
+ virtual bool eval_f(Index n, const Number* x, bool new_x, Number& obj_value);
+
+ /** Method to return the gradient of the objective */
+ virtual bool eval_grad_f(Index n, const Number* x, bool new_x, Number* grad_f);
+
+ /** Method to return the constraint residuals */
+ virtual bool eval_g(Index n, const Number* x, bool new_x, Index m, Number* g);
+
+ /** Method to return:
+ * 1) The structure of the jacobian (if "values" is NULL)
+ * 2) The values of the jacobian (if "values" is not NULL)
+ */
+ virtual bool eval_jac_g(Index n, const Number* x, bool new_x,Index m, Index nele_jac, Index* iRow, Index *jCol,Number* values);
+
+ /** Method to return:
+ * 1) The structure of the hessian of the lagrangian (if "values" is NULL)
+ * 2) The values of the hessian of the lagrangian (if "values" is not NULL)
+ */
+ virtual bool eval_h(Index n, const Number* x, bool new_x,Number obj_factor, Index m, const Number* lambda,bool new_lambda, Index nele_hess, Index* iRow,Index* jCol, Number* values);
+
+ /** This method is called when the algorithm is complete so the TNLP can store/write the solution */
+ virtual void finalize_solution(SolverReturn status,Index n, const Number* x, Number obj_value);
+
+ virtual const SosInfo * sosConstraints() const{return NULL;}
+ virtual const BranchingInfo* branchingInfo() const{return NULL;}
+
+ const double * getX(); //Returns a pointer to a matrix of size of 1*numVars_
+ //with final value for the primal variables.
+
+ const double * getGrad(); //Returns a pointer to a matrix of size of 1*numVars_
+ //with final value of gradient for the primal variables.
+
+ const double * getHess(); //Returns a pointer to a matrix of size of numVars_*numVars_
+ //with final value of hessian for the primal variables.
+
+ double getObjVal(); //Returns the output of the final value of the objective.
+
+ double iterCount(); //Returns the iteration count
+
+ int returnStatus(); //Returns the status count
+};
+
+#endif
diff --git a/sci_gateway/cpp/minuncTMINLP.hpp b/sci_gateway/cpp/minuncTMINLP.hpp
new file mode 100644
index 0000000..2b6e954
--- /dev/null
+++ b/sci_gateway/cpp/minuncTMINLP.hpp
@@ -0,0 +1,113 @@
+// Copyright (C) 2016 - IIT Bombay - FOSSEE
+//
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+// Author: Harpreet Singh, Pranav Deshpande and Akshay Miterani
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+
+#define __USE_DEPRECATED_STACK_FUNCTIONS__
+#ifndef minuncTMINLP_HPP
+#define minuncTMINLP_HPP
+
+#include "BonTMINLP.hpp"
+#include "IpTNLP.hpp"
+#include "call_scilab.h"
+
+using namespace Ipopt;
+using namespace Bonmin;
+
+class minuncTMINLP : public TMINLP
+{
+ private:
+
+ Index numVars_; //Number of input variables
+
+ Index intconSize_;
+
+ const Number *varGuess_= NULL; //varGuess_ is a pointer to a matrix of size of 1*numVars_ with initial guess of all variables.
+
+ Number *finalX_= NULL; //finalX_ is a pointer to a matrix of size of 1*numVars_ with final value for the primal variables.
+
+ Number finalObjVal_; //finalObjVal_ is a scalar with the final value of the objective.
+
+ Number *intcon_ = NULL;
+
+ int status_; //Solver return status
+ minuncTMINLP(const minuncTMINLP&);
+ minuncTMINLP& operator=(const minuncTMINLP&);
+
+public:
+ // Constructor
+ minuncTMINLP(Index nV, Number *x0, Index intconSize, Number *intcon):numVars_(nV),varGuess_(x0),intconSize_(intconSize),intcon_(intcon),finalX_(0),finalObjVal_(1e20){ }
+
+ /** default destructor */
+ virtual ~minuncTMINLP();
+
+ virtual bool get_variables_types(Index n, VariableType* var_types);
+
+ virtual bool get_variables_linearity(Index n, Ipopt::TNLP::LinearityType* var_types);
+
+ virtual bool get_constraints_linearity(Index m, Ipopt::TNLP::LinearityType* const_types);
+
+ /** Method to return some info about the nlp */
+ virtual bool get_nlp_info(Index& n, Index& m, Index& nnz_jac_g,
+ Index& nnz_h_lag, TNLP::IndexStyleEnum& index_style);
+
+ /** Method to return the bounds for my problem */
+ virtual bool get_bounds_info(Index n, Number* x_l, Number* x_u,
+ Index m, Number* g_l, Number* g_u);
+
+ /** Method to return the starting point for the algorithm */
+ virtual bool get_starting_point(Index n, bool init_x, Number* x,
+ bool init_z, Number* z_L, Number* z_U,
+ Index m, bool init_lambda,
+ Number* lambda);
+
+ /** Method to return the objective value */
+ virtual bool eval_f(Index n, const Number* x, bool new_x, Number& obj_value);
+
+ /** Method to return the gradient of the objective */
+ virtual bool eval_grad_f(Index n, const Number* x, bool new_x, Number* grad_f);
+
+ /** Method to return the constraint residuals */
+ virtual bool eval_g(Index n, const Number* x, bool new_x, Index m, Number* g);
+
+ /** Method to return:
+ * 1) The structure of the jacobian (if "values" is NULL)
+ * 2) The values of the jacobian (if "values" is not NULL)
+ */
+ virtual bool eval_jac_g(Index n, const Number* x, bool new_x,Index m, Index nele_jac, Index* iRow, Index *jCol,Number* values);
+
+ /** Method to return:
+ * 1) The structure of the hessian of the lagrangian (if "values" is NULL)
+ * 2) The values of the hessian of the lagrangian (if "values" is not NULL)
+ */
+ virtual bool eval_h(Index n, const Number* x, bool new_x,Number obj_factor, Index m, const Number* lambda,bool new_lambda, Index nele_hess, Index* iRow,Index* jCol, Number* values);
+
+ /** This method is called when the algorithm is complete so the TNLP can store/write the solution */
+ virtual void finalize_solution(SolverReturn status,Index n, const Number* x, Number obj_value);
+
+ virtual const SosInfo * sosConstraints() const{return NULL;}
+ virtual const BranchingInfo* branchingInfo() const{return NULL;}
+
+ const double * getX(); //Returns a pointer to a matrix of size of 1*numVars_
+ //with final value for the primal variables.
+
+ const double * getGrad(); //Returns a pointer to a matrix of size of 1*numVars_
+ //with final value of gradient for the primal variables.
+
+ const double * getHess(); //Returns a pointer to a matrix of size of numVars_*numVars_
+ //with final value of hessian for the primal variables.
+
+ double getObjVal(); //Returns the output of the final value of the objective.
+
+ double iterCount(); //Returns the iteration count
+
+ int returnStatus(); //Returns the status count
+};
+
+#endif
diff --git a/sci_gateway/cpp/sci_QuadTMINLP.cpp b/sci_gateway/cpp/sci_QuadTMINLP.cpp
new file mode 100644
index 0000000..a424b47
--- /dev/null
+++ b/sci_gateway/cpp/sci_QuadTMINLP.cpp
@@ -0,0 +1,230 @@
+// Copyright (C) 2016 - IIT Bombay - FOSSEE
+//
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+// Author: Harpreet Singh, Pranav Deshpande and Akshay Miterani
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+
+#include "QuadTMINLP.hpp"
+#include "IpIpoptData.hpp"
+
+extern "C"{
+#include <sciprint.h>
+}
+
+// Go to http://coin-or.org/Ipopt and http://coin-or.org/Bonmin for the details of the below methods
+
+// Set the type of every variable - CONTINUOUS or INTEGER
+bool QuadTMINLP::get_variables_types(Index n, VariableType* var_types)
+{
+ n = numVars_;
+ for(int i=0; i < n; i++)
+ var_types[i] = CONTINUOUS;
+ for(int i=0 ; i < intconSize_ ; ++i)
+ var_types[(int)(intcon_[i]-1)] = INTEGER;
+ return true;
+}
+
+// The linearity of the variables - LINEAR or NON_LINEAR
+bool QuadTMINLP::get_variables_linearity(Index n, Ipopt::TNLP::LinearityType* var_types)
+{
+ /*
+ n = numVars_;
+ for(int i = 0; i < n; i++)
+ var_types[i] = Ipopt::TNLP::LINEAR;
+ */
+ return true;
+}
+
+// The linearity of the constraints - LINEAR or NON_LINEAR
+bool QuadTMINLP::get_constraints_linearity(Index m, Ipopt::TNLP::LinearityType* const_types)
+{
+ m = numCons_;
+ for(int i = 0; i < m; i++)
+ const_types[i] = Ipopt::TNLP::LINEAR;
+ return true;
+}
+
+// Get MINLP info such as the number of variables,constraints,no.of elements in jacobian and hessian to allocate memory
+bool QuadTMINLP::get_nlp_info(Index& n, Index&m, Index& nnz_jac_g, Index& nnz_h_lag, TNLP::IndexStyleEnum& index_style)
+{
+ n=numVars_; // Number of variables
+ m=numCons_; // Number of constraints
+ nnz_jac_g = n*m; // No. of elements in Jacobian of constraints
+ nnz_h_lag = (n*(n+1))/2; // No. of elements in lower traingle of Hessian of the Lagrangian.
+ index_style=TNLP::C_STYLE; // Index style of matrices
+ return true;
+}
+
+// Get the variables and constraints bound info
+bool QuadTMINLP::get_bounds_info(Index n, Number* x_l, Number* x_u, Index m, Number* g_l, Number* g_u)
+{
+ unsigned int i;
+ for(i=0;i<n;i++){
+ x_l[i]=varLB_[i];
+ x_u[i]=varUB_[i];
+ }
+
+ for(i=0;i<m;i++){
+ g_l[i]=conLB_[i];
+ g_u[i]=conUB_[i];
+ }
+ return true;
+}
+
+// This method sets initial values for all the required vectors. We take 0 by default.
+bool QuadTMINLP::get_starting_point(Index n, bool init_x, Number* x,
+ bool init_z, Number* z_L, Number* z_U,
+ Index m, bool init_lambda,
+ Number* lambda)
+{
+ if (init_x == true){ //we need to set initial values for vector x
+ for (Index var=0;var<n;var++)
+ x[var]=varGuess_[var];//initialize with 0 or we can change.
+ }
+
+ if (init_z == true){ //we need to provide initial values for vector bound multipliers
+ for (Index var=0;var<n;++var){
+ z_L[var]=0.0; //initialize with 0 or we can change.
+ z_U[var]=0.0;//initialize with 0 or we can change.
+ }
+ }
+
+ if (init_lambda == true){ //we need to provide initial values for lambda values.
+ for (Index var=0;var<m;++var){
+ lambda[var]=0.0; //initialize with 0 or we can change.
+ }
+ }
+
+ return true;
+}
+
+// Evaluate the objective function at a point
+bool QuadTMINLP::eval_f(Index n, const Number* x, bool new_x, Number& obj_value)
+{
+ unsigned int i,j;
+ obj_value=0;
+ for (i=0;i<n;i++){
+ for (j=0;j<n;j++){
+ obj_value+=0.5*x[i]*x[j]*qMatrix_[n*i+j];
+ }
+ obj_value+=x[i]*lMatrix_[i];
+ }
+ return true;
+}
+
+// Get the value of gradient of objective function at vector x.
+bool QuadTMINLP::eval_grad_f(Index n, const Number* x, bool new_x, Number* grad_f)
+{
+ unsigned int i,j;
+ for(i=0;i<n;i++)
+ {
+ grad_f[i]=lMatrix_[i];
+ for(j=0;j<n;j++)
+ {
+ grad_f[i]+=(qMatrix_[n*i+j])*x[j];
+ }
+ }
+ return true;
+}
+
+// Get the values of constraints at vector x.
+bool QuadTMINLP::eval_g(Index n, const Number* x, bool new_x, Index m, Number* g)
+{
+ unsigned int i,j;
+ for(i=0;i<m;i++)
+ {
+ g[i]=0;
+ for(j=0;j<n;j++)
+ {
+ g[i]+=x[j]*conMatrix_[i+j*m];
+ }
+ }
+ return true;
+}
+
+// The Jacobian Matrix
+bool QuadTMINLP::eval_jac_g(Index n, const Number* x, bool new_x,
+ Index m, Index nnz_jac, Index* iRow, Index *jCol,
+ Number* values)
+{
+ //It asks for the structure of the jacobian.
+ if (values==NULL){ //Structure of jacobian (full structure)
+ int index=0;
+ for (int var=0;var<m;++var)//no. of constraints
+ for (int flag=0;flag<n;++flag){//no. of variables
+ iRow[index]=var;
+ jCol[index]=flag;
+ index++;
+ }
+ }
+ //It asks for values
+ else {
+ int index=0;
+ for (int var=0;var<m;++var)
+ for (int flag=0;flag<n;++flag)
+ values[index++]=conMatrix_[var+flag*m];
+ }
+ return true;
+}
+
+/*
+The structure of the Hessain matrix and the values
+*/
+bool QuadTMINLP::eval_h(Index n, const Number* x, bool new_x,
+ Number obj_factor, Index m, const Number* lambda,
+ bool new_lambda, Index nele_hess, Index* iRow,
+ Index* jCol, Number* values)
+{
+ if (values==NULL){
+ Index idx=0;
+ for (Index row = 0; row < n; row++) {
+ for (Index col = 0; col <= row; col++) {
+ iRow[idx] = row;
+ jCol[idx] = col;
+ idx++;
+ }
+ }
+ }
+ else {
+ Index index=0;
+ for (Index row=0;row < n;++row){
+ for (Index col=0; col <= row; ++col){
+ values[index++]=obj_factor*(qMatrix_[n*row+col]);
+ }
+ }
+ }
+ return true;
+}
+
+void QuadTMINLP::finalize_solution(TMINLP::SolverReturn status, Index n, const Number* x,Number obj_value)
+{
+
+ finalX_ = (double*)malloc(sizeof(double) * numVars_ * 1);
+ for (Index i=0; i<n; i++)
+ {
+ finalX_[i] = x[i];
+ }
+
+ finalObjVal_ = obj_value;
+ status_ = status;
+}
+
+const double * QuadTMINLP::getX()
+{
+ return finalX_;
+}
+
+double QuadTMINLP::getObjVal()
+{
+ return finalObjVal_;
+}
+
+int QuadTMINLP::returnStatus()
+{
+ return status_;
+}
diff --git a/sci_gateway/cpp/sci_iofunc.cpp b/sci_gateway/cpp/sci_iofunc.cpp
new file mode 100644
index 0000000..f05839c
--- /dev/null
+++ b/sci_gateway/cpp/sci_iofunc.cpp
@@ -0,0 +1,333 @@
+// Symphony Toolbox for Scilab
+// (Definition of) Functions for input and output from Scilab
+// By Keyur Joshi
+
+#include "api_scilab.h"
+#include "Scierror.h"
+#include "sciprint.h"
+#include "BOOL.h"
+#include <localization.h>
+#include "call_scilab.h"
+#include <string.h>
+
+
+using namespace std;
+
+int getDoubleFromScilab(int argNum, double *dest)
+{
+ //data declarations
+ SciErr sciErr;
+ int iRet,*varAddress;
+ const char errMsg[]="Wrong type for input argument #%d: A double is expected.\n";
+ const int errNum=999;
+ //get variable address
+ sciErr = getVarAddressFromPosition(pvApiCtx, argNum, &varAddress);
+ if (sciErr.iErr)
+ {
+ printError(&sciErr, 0);
+ return 1;
+ }
+ //check that it is a non-complex double
+ if ( !isDoubleType(pvApiCtx,varAddress) || isVarComplex(pvApiCtx,varAddress) )
+ {
+ Scierror(errNum,errMsg,argNum);
+ return 1;
+ }
+ //retrieve and store
+ iRet = getScalarDouble(pvApiCtx, varAddress, dest);
+ if(iRet)
+ {
+ Scierror(errNum,errMsg,argNum);
+ return 1;
+ }
+ return 0;
+}
+
+int getUIntFromScilab(int argNum, int *dest)
+{
+ SciErr sciErr;
+ int iRet,*varAddress;
+ double inputDouble;
+ const char errMsg[]="Wrong type for input argument #%d: A nonnegative integer is expected.\n";
+ const int errNum=999;
+ //same steps as above
+ sciErr = getVarAddressFromPosition(pvApiCtx, argNum, &varAddress);
+ if (sciErr.iErr)
+ {
+ printError(&sciErr, 0);
+ return 1;
+ }
+ if ( !isDoubleType(pvApiCtx,varAddress) || isVarComplex(pvApiCtx,varAddress) )
+ {
+ Scierror(errNum,errMsg,argNum);
+ return 1;
+ }
+ iRet = getScalarDouble(pvApiCtx, varAddress, &inputDouble);
+ //check that an unsigned int is stored in the double by casting and recasting
+ if(iRet || ((double)((unsigned int)inputDouble))!=inputDouble)
+ {
+ Scierror(errNum,errMsg,argNum);
+ return 1;
+ }
+ *dest=(unsigned int)inputDouble;
+ return 0;
+}
+
+int getIntFromScilab(int argNum, int *dest)
+{
+ SciErr sciErr;
+ int iRet,*varAddress;
+ double inputDouble;
+ const char errMsg[]="Wrong type for input argument #%d: An integer is expected.\n";
+ const int errNum=999;
+ //same steps as above
+ sciErr = getVarAddressFromPosition(pvApiCtx, argNum, &varAddress);
+ if (sciErr.iErr)
+ {
+ printError(&sciErr, 0);
+ return 1;
+ }
+ if ( !isDoubleType(pvApiCtx,varAddress) || isVarComplex(pvApiCtx,varAddress) )
+ {
+ Scierror(errNum,errMsg,argNum);
+ return 1;
+ }
+ iRet = getScalarDouble(pvApiCtx, varAddress, &inputDouble);
+ //check that an int is stored in the double by casting and recasting
+ if(iRet || ((double)((int)inputDouble))!=inputDouble)
+ {
+ Scierror(errNum,errMsg,argNum);
+ return 1;
+ }
+ *dest=(int)inputDouble;
+ return 0;
+}
+
+int getFixedSizeDoubleMatrixFromScilab(int argNum, int rows, int cols, double **dest)
+{
+ int *varAddress,inputMatrixRows,inputMatrixCols;
+ SciErr sciErr;
+ const char errMsg[]="Wrong type for input argument #%d: A matrix of double of size %d by %d is expected.\n";
+ const int errNum=999;
+ //same steps as above
+ sciErr = getVarAddressFromPosition(pvApiCtx, argNum, &varAddress);
+ if (sciErr.iErr)
+ {
+ printError(&sciErr, 0);
+ return 1;
+ }
+ if ( !isDoubleType(pvApiCtx,varAddress) || isVarComplex(pvApiCtx,varAddress) )
+ {
+ Scierror(errNum,errMsg,argNum,rows,cols);
+ return 1;
+ }
+ sciErr = getMatrixOfDouble(pvApiCtx, varAddress, &inputMatrixRows, &inputMatrixCols,NULL);
+ if (sciErr.iErr)
+ {
+ printError(&sciErr, 0);
+ return 1;
+ }
+ //check that the matrix has the correct number of rows and columns
+ if(inputMatrixRows!=rows || inputMatrixCols!=cols)
+ {
+ Scierror(errNum,errMsg,argNum,rows,cols);
+ return 1;
+ }
+ getMatrixOfDouble(pvApiCtx, varAddress, &inputMatrixRows, &inputMatrixCols, dest);
+ return 0;
+}
+
+int getDoubleMatrixFromScilab(int argNum, int *rows, int *cols, double **dest)
+{
+ int *varAddress;
+ SciErr sciErr;
+ const char errMsg[]="Wrong type for input argument #%d: A matrix of double is expected.\n";
+ const int errNum=999;
+ //same steps as above
+ sciErr = getVarAddressFromPosition(pvApiCtx, argNum, &varAddress);
+ if (sciErr.iErr)
+ {
+ printError(&sciErr, 0);
+ return 1;
+ }
+ if ( !isDoubleType(pvApiCtx,varAddress) || isVarComplex(pvApiCtx,varAddress) )
+ {
+ Scierror(errNum,errMsg,argNum);
+ return 1;
+ }
+ getMatrixOfDouble(pvApiCtx, varAddress, rows, cols, dest);
+ if (sciErr.iErr)
+ {
+ printError(&sciErr, 0);
+ return 1;
+ }
+ return 0;
+}
+
+int getFixedSizeDoubleMatrixInList(int argNum, int itemPos, int rows, int cols, double **dest)
+{
+ int *varAddress,inputMatrixRows,inputMatrixCols;
+ SciErr sciErr;
+ const char errMsg[]="Wrong type for input argument #%d: A matrix of double of size %d by %d is expected.\n";
+ const int errNum=999;
+ //same steps as above
+ sciErr = getVarAddressFromPosition(pvApiCtx, argNum, &varAddress);
+ if (sciErr.iErr)
+ {
+ printError(&sciErr, 0);
+ return 1;
+ }
+
+ getMatrixOfDoubleInList(pvApiCtx, varAddress, itemPos, &rows, &cols, dest);
+ if (sciErr.iErr)
+ {
+ printError(&sciErr, 0);
+ return 1;
+ }
+ return 0;
+}
+
+int getStringFromScilab(int argNum,char **dest)
+{
+ int *varAddress,inputMatrixRows,inputMatrixCols;
+ SciErr sciErr;
+ sciErr = getVarAddressFromPosition(pvApiCtx, argNum, &varAddress);
+
+ //check whether there is an error or not.
+ if (sciErr.iErr)
+ {
+ printError(&sciErr, 0);
+ return 1;
+ }
+ if ( !isStringType(pvApiCtx,varAddress) )
+ {
+ Scierror(999,"Wrong type for input argument 1: A file name is expected.\n");
+ return 1;
+ }
+ //read the value in that pointer pointing to file name
+ getAllocatedSingleString(pvApiCtx, varAddress, dest);
+
+}
+
+bool getFunctionFromScilab(int n,char name[], double *x,int posFirstElementOnStackForSF,int nOfRhsOnSF,int nOfLhsOnSF, double **dest)
+{
+ double check;
+ createMatrixOfDouble(pvApiCtx, posFirstElementOnStackForSF, 1, n, x);
+ C2F(scistring)(&posFirstElementOnStackForSF,name,&nOfLhsOnSF,&nOfRhsOnSF,(unsigned long)strlen(name));
+
+ if(getDoubleFromScilab(posFirstElementOnStackForSF+1,&check))
+ {
+ return true;
+ }
+ if (check==1)
+ {
+ return true;
+ }
+ else
+ {
+ int x_rows, x_cols;
+ if(getDoubleMatrixFromScilab(posFirstElementOnStackForSF, &x_rows, &x_cols, dest))
+ {
+ sciprint("No results ");
+ return true;
+
+ }
+ }
+ return 0;
+}
+
+bool getHessFromScilab(int n,int numConstr_,char name[], double *x,double *obj,double *lambda,int posFirstElementOnStackForSF,int nOfRhsOnSF,int nOfLhsOnSF, double **dest)
+{
+ double check;
+ createMatrixOfDouble(pvApiCtx, posFirstElementOnStackForSF, 1, n, x);
+ createMatrixOfDouble(pvApiCtx, posFirstElementOnStackForSF+1, 1, 1, obj);
+ createMatrixOfDouble(pvApiCtx, posFirstElementOnStackForSF+2, 1, numConstr_, lambda);
+ C2F(scistring)(&posFirstElementOnStackForSF,name,&nOfLhsOnSF,&nOfRhsOnSF,(unsigned long)strlen(name));
+
+ if(getDoubleFromScilab(posFirstElementOnStackForSF+1,&check))
+ {
+ return true;
+ }
+ if (check==1)
+ {
+ return true;
+ }
+ else
+ {
+ int x_rows, x_cols;
+ if(getDoubleMatrixFromScilab(posFirstElementOnStackForSF, &x_rows, &x_cols, dest))
+ {
+ sciprint("No results ");
+ return 1;
+ }
+ }
+ return 0;
+}
+
+int return0toScilab()
+{
+ int iRet;
+ //create variable in scilab
+ iRet = createScalarDouble(pvApiCtx, nbInputArgument(pvApiCtx)+1,0);
+ if(iRet)
+ {
+ /* If error, no return variable */
+ AssignOutputVariable(pvApiCtx, 1) = 0;
+ return 1;
+ }
+ //make it the output variable
+ AssignOutputVariable(pvApiCtx, 1) = nbInputArgument(pvApiCtx)+1;
+ //return it to scilab
+ //ReturnArguments(pvApiCtx);
+ return 0;
+}
+
+int returnDoubleToScilab(double retVal)
+{
+ int iRet;
+ //same steps as above
+ iRet = createScalarDouble(pvApiCtx, nbInputArgument(pvApiCtx)+1,retVal);
+ if(iRet)
+ {
+ /* If error, no return variable */
+ AssignOutputVariable(pvApiCtx, 1) = 0;
+ return 1;
+ }
+ AssignOutputVariable(pvApiCtx, 1) = nbInputArgument(pvApiCtx)+1;
+ //ReturnArguments(pvApiCtx);
+ return 0;
+}
+
+int returnDoubleMatrixToScilab(int itemPos, int rows, int cols, double *dest)
+{
+ SciErr sciErr;
+ //same steps as above
+ sciErr = createMatrixOfDouble(pvApiCtx, nbInputArgument(pvApiCtx) + itemPos, rows, cols, dest);
+ if (sciErr.iErr)
+ {
+ printError(&sciErr, 0);
+ return 1;
+ }
+
+ AssignOutputVariable(pvApiCtx, itemPos) = nbInputArgument(pvApiCtx)+itemPos;
+
+ return 0;
+}
+
+int returnIntegerMatrixToScilab(int itemPos, int rows, int cols, int *dest)
+{
+ SciErr sciErr;
+ //same steps as above
+ sciErr = createMatrixOfInteger32(pvApiCtx, nbInputArgument(pvApiCtx) + itemPos, rows, cols, dest);
+ if (sciErr.iErr)
+ {
+ printError(&sciErr, 0);
+ return 1;
+ }
+
+ AssignOutputVariable(pvApiCtx, itemPos) = nbInputArgument(pvApiCtx)+itemPos;
+
+ return 0;
+}
+
+
diff --git a/sci_gateway/cpp/sci_iofunc.hpp b/sci_gateway/cpp/sci_iofunc.hpp
new file mode 100644
index 0000000..7e18951
--- /dev/null
+++ b/sci_gateway/cpp/sci_iofunc.hpp
@@ -0,0 +1,25 @@
+// Symphony Toolbox for Scilab
+// (Declaration of) Functions for input and output from Scilab
+// By Keyur Joshi
+
+#ifndef SCI_IOFUNCHEADER
+#define SCI_IOFUNCHEADER
+
+//input
+int getDoubleFromScilab(int argNum, double *dest);
+int getUIntFromScilab(int argNum, int *dest);
+int getIntFromScilab(int argNum, int *dest);
+int getFixedSizeDoubleMatrixFromScilab(int argNum, int rows, int cols, double **dest);
+int getDoubleMatrixFromScilab(int argNum, int *rows, int *cols, double **dest);
+int getFixedSizeDoubleMatrixInList(int argNum, int itemPos, int rows, int cols, double **dest);
+int getStringFromScilab(int argNum,char** dest);
+bool getFunctionFromScilab(int n,char name[], double *x,int posFirstElementOnStackForSF,int nOfRhsOnSF,int nOfLhsOnSF, double **dest);
+bool getHessFromScilab(int n,int numConstr_,char name[], double *x,double *obj,double *lambda,int posFirstElementOnStackForSF,int nOfRhsOnSF,int nOfLhsOnSF, double **dest);
+
+//output
+int return0toScilab();
+int returnDoubleToScilab(double retVal);
+int returnDoubleMatrixToScilab(int itemPos, int rows, int cols, double *dest);
+int returnIntegerMatrixToScilab(int itemPos, int rows, int cols, int *dest);
+
+#endif //SCI_IOFUNCHEADER
diff --git a/sci_gateway/cpp/sci_minbndTMINLP.cpp b/sci_gateway/cpp/sci_minbndTMINLP.cpp
new file mode 100644
index 0000000..f26c089
--- /dev/null
+++ b/sci_gateway/cpp/sci_minbndTMINLP.cpp
@@ -0,0 +1,218 @@
+// Copyright (C) 2015 - IIT Bombay - FOSSEE
+//
+// Author: Harpreet Singh
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+
+#include "minbndTMINLP.hpp"
+#include "sci_iofunc.hpp"
+
+extern "C"
+{
+#include "call_scilab.h"
+#include <api_scilab.h>
+#include <Scierror.h>
+#include <BOOL.h>
+#include <localization.h>
+#include <sciprint.h>
+#include <string.h>
+#include <assert.h>
+}
+
+using namespace Ipopt;
+using namespace Bonmin;
+
+minbndTMINLP::~minbndTMINLP()
+{
+ if(finalX_) delete[] finalX_;
+}
+
+// Set the type of every variable - CONTINUOUS or INTEGER
+bool minbndTMINLP::get_variables_types(Index n, VariableType* var_types)
+{
+ n = numVars_;
+ for(int i=0; i < n; i++)
+ var_types[i] = CONTINUOUS;
+ for(int i=0 ; i < intconSize_ ; ++i)
+ var_types[(int)(intcon_[i]-1)] = INTEGER;
+ return true;
+}
+
+// The linearity of the variables - LINEAR or NON_LINEAR
+bool minbndTMINLP::get_variables_linearity(Index n, Ipopt::TNLP::LinearityType* var_types)
+{ return true; }
+
+// The linearity of the constraints - LINEAR or NON_LINEAR
+bool minbndTMINLP::get_constraints_linearity(Index m, Ipopt::TNLP::LinearityType* const_types)
+{ return true;}
+
+//get NLP info such as number of variables,constraints,no.of elements in jacobian and hessian to allocate memory
+bool minbndTMINLP::get_nlp_info(Index& n, Index& m, Index& nnz_jac_g, Index& nnz_h_lag, TNLP::IndexStyleEnum& index_style)
+{
+ n=numVars_; // Number of variables
+ m=0; // Number of constraints
+ nnz_jac_g = 0; // No. of elements in Jacobian of constraints
+ nnz_h_lag = n*(n+1)/2; // No. of elements in lower traingle of Hessian of the Lagrangian.
+ index_style=TNLP::C_STYLE; // Index style of matrices
+ return true;
+}
+
+//get variable and constraint bound info
+bool minbndTMINLP::get_bounds_info(Index n, Number* x_l, Number* x_u, Index m, Number* g_l, Number* g_u)
+{
+ unsigned int i;
+ for(i=0;i<n;i++)
+ {
+ x_l[i]=lb_[i]+0.0000001;
+ x_u[i]=ub_[i]-0.0000001;
+ }
+
+ g_l=NULL;
+ g_u=NULL;
+ return true;
+}
+
+// return the value of the constraints: g(x)
+bool minbndTMINLP::eval_g(Index n, const Number* x, bool new_x, Index m, Number* g)
+{
+ // return the value of the constraints: g(x)
+ g=NULL;
+ return true;
+}
+
+// return the structure or values of the jacobian
+bool minbndTMINLP::eval_jac_g(Index n, const Number* x, bool new_x,Index m, Index nele_jac, Index* iRow, Index *jCol,Number* values)
+{
+ if (values == NULL)
+ {
+ // return the structure of the jacobian of the constraints
+ iRow=NULL;
+ jCol=NULL;
+ }
+ else
+ {
+ values=NULL;
+ }
+ return true;
+}
+
+//get value of objective function at vector x
+bool minbndTMINLP::eval_f(Index n, const Number* x, bool new_x, Number& obj_value)
+{
+ char name[20]="_f";
+ Number *obj;
+ if (getFunctionFromScilab(n,name,x, 7, 1,2,&obj))
+ {
+ return false;
+ }
+ obj_value = *obj;
+ return true;
+}
+
+//get value of gradient of objective function at vector x.
+bool minbndTMINLP::eval_grad_f(Index n, const Number* x, bool new_x, Number* grad_f)
+{
+ char name[20]="_gradf";
+ Number *resg;
+ if (getFunctionFromScilab(n,name,x, 7, 1,2,&resg))
+ {
+ return false;
+ }
+
+ Index i;
+ for(i=0;i<numVars_;i++)
+ {
+ grad_f[i]=resg[i];
+ }
+ return true;
+}
+
+// This method sets initial values for required vectors . For now we are assuming 0 to all values.
+bool minbndTMINLP::get_starting_point(Index n, bool init_x, Number* x,bool init_z, Number* z_L, Number* z_U,Index m, bool init_lambda,Number* lambda)
+{
+ assert(init_x == true);
+ assert(init_z == false);
+ assert(init_lambda == false);
+ if (init_x == true)
+ { //we need to set initial values for vector x
+ for (Index var=0;var<n;var++)
+ {x[var]=0.0;}//initialize with 0.
+ }
+ return true;
+}
+
+/*
+ * Return either the sparsity structure of the Hessian of the Lagrangian,
+ * or the values of the Hessian of the Lagrangian for the given values for
+ * x,lambda,obj_factor.
+*/
+
+bool minbndTMINLP::eval_h(Index n, const Number* x, bool new_x,Number obj_factor, Index m, const Number* lambda,bool new_lambda, Index nele_hess, Index* iRow,Index* jCol, Number* values)
+{
+ double check;
+ if (values==NULL)
+ {
+ Index idx=0;
+ for (Index row = 0; row < numVars_; row++)
+ {
+ for (Index col = 0; col <= row; col++)
+ { iRow[idx] = row;
+ jCol[idx] = col;
+ idx++;
+ }
+ }
+ }
+
+ else
+ { char name[20]="_gradhess";
+ Number *resh;
+ if (getFunctionFromScilab(n,name,x, 7, 1,2,&resh))
+ {
+ return false;
+ }
+ Index index=0;
+ for (Index row=0;row < numVars_ ;++row)
+ {
+ for (Index col=0; col <= row; ++col)
+ {
+ values[index++]=obj_factor*(resh[numVars_*row+col]);
+ }
+ }
+ }
+ return true;
+}
+
+void minbndTMINLP::finalize_solution(SolverReturn status,Index n, const Number* x, Number obj_value)
+{
+ finalObjVal_ = obj_value;
+ status_ = status;
+ if(status==0 ||status== 3)
+ {
+ finalX_ = new double[n];
+ for (Index i=0; i<numVars_; i++)
+ {
+ finalX_[i] = x[i];
+ }
+ }
+
+}
+
+const double * minbndTMINLP::getX()
+{
+ return finalX_;
+}
+
+double minbndTMINLP::getObjVal()
+{
+ return finalObjVal_;
+}
+
+int minbndTMINLP::returnStatus()
+{
+ return status_;
+}
diff --git a/sci_gateway/cpp/sci_minconTMINLP.cpp b/sci_gateway/cpp/sci_minconTMINLP.cpp
new file mode 100644
index 0000000..350594d
--- /dev/null
+++ b/sci_gateway/cpp/sci_minconTMINLP.cpp
@@ -0,0 +1,324 @@
+// Copyright (C) 2015 - IIT Bombay - FOSSEE
+//
+// Author: Harpreet Singh, Pranav Deshpande and Akshay Miterani
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+
+#include "minconTMINLP.hpp"
+#include "sci_iofunc.hpp"
+
+extern "C"
+{
+#include "call_scilab.h"
+#include <api_scilab.h>
+#include <Scierror.h>
+#include <BOOL.h>
+#include <localization.h>
+#include <sciprint.h>
+#include <string.h>
+#include <assert.h>
+}
+
+using namespace Ipopt;
+using namespace Bonmin;
+
+//#define DEBUG 0
+
+minconTMINLP::~minconTMINLP()
+{
+ if(finalX_) delete[] finalX_;
+}
+
+// Set the type of every variable - CONTINUOUS or INTEGER
+bool minconTMINLP::get_variables_types(Index n, VariableType* var_types)
+{
+ #ifdef DEBUG
+ sciprint("Code is in get_variables_types\n");
+ #endif
+ n = numVars_;
+ for(int i=0; i < n; i++)
+ var_types[i] = CONTINUOUS;
+ for(int i=0 ; i < intconSize_ ; ++i)
+ var_types[(int)(intcon_[i]-1)] = INTEGER;
+ return true;
+}
+
+// The linearity of the variables - LINEAR or NON_LINEAR
+bool minconTMINLP::get_variables_linearity(Index n, Ipopt::TNLP::LinearityType* var_types)
+{
+ #ifdef DEBUG
+ sciprint("Code is in get_variables_linearity\n");
+ #endif
+ for(int i=0;i<n;i++)
+ {
+ var_types[i] = Ipopt::TNLP::NON_LINEAR;
+ }
+ return true; }
+
+// The linearity of the constraints - LINEAR or NON_LINEAR
+bool minconTMINLP::get_constraints_linearity(Index m, Ipopt::TNLP::LinearityType* const_types)
+{
+
+ #ifdef DEBUG
+ sciprint("Code is in get_constraints_linearity\n");
+ #endif
+ for(int i=0;i<numLC_;i++)
+ {
+ const_types[i] = Ipopt::TNLP::LINEAR;
+ }
+
+ for(int i=numLC_;i<m;i++)
+ {
+ const_types[i] = Ipopt::TNLP::NON_LINEAR;
+ }
+ return true;}
+
+//get NLP info such as number of variables,constraints,no.of elements in jacobian and hessian to allocate memory
+bool minconTMINLP::get_nlp_info(Index& n, Index& m, Index& nnz_jac_g, Index& nnz_h_lag, TNLP::IndexStyleEnum& index_style)
+{
+ #ifdef DEBUG
+ sciprint("Code is in get_nlp_info\n");
+ #endif
+ n=numVars_; // Number of variables
+ m=numCons_; // Number of constraints
+ nnz_jac_g = n*m; // No. of elements in Jacobian of constraints
+ nnz_h_lag = n*n; // No. of elements in Hessian of the Lagrangian.
+ index_style=TNLP::C_STYLE; // Index style of matrices
+ return true;
+}
+
+//get variable and constraint bound info
+bool minconTMINLP::get_bounds_info(Index n, Number* x_l, Number* x_u, Index m, Number* g_l, Number* g_u)
+{
+ #ifdef DEBUG
+ sciprint("Code is in get_bounds_info\n");
+ #endif
+ unsigned int i;
+ for(i=0;i<n;i++)
+ {
+ x_l[i]=lb_[i];
+ x_u[i]=ub_[i];
+ }
+ for(i=0;i<m;i++)
+ {
+ g_l[i]=conLb_[i];
+ g_u[i]=conUb_[i];
+ }
+ return true;
+}
+
+// This method sets initial values for required vectors . For now we are assuming 0 to all values.
+bool minconTMINLP::get_starting_point(Index n, bool init_x, Number* x,bool init_z, Number* z_L, Number* z_U,Index m, bool init_lambda,Number* lambda)
+{
+ assert(init_x == true);
+ assert(init_z == false);
+ assert(init_lambda == false);
+ if (init_x == true)
+ { //we need to set initial values for vector x
+ for (Index var=0;var<n;var++)
+ {x[var]=x0_[var];}//initialize with 0.
+ }
+ return true;
+}
+
+//get value of objective function at vector x
+bool minconTMINLP::eval_f(Index n, const Number* x, bool new_x, Number& obj_value)
+{
+ #ifdef DEBUG
+ sciprint("Code is eval_f\n");
+ #endif
+ char name[20]="_f";
+ Number *obj;
+ if (getFunctionFromScilab(n,name,x, 7, 1,2,&obj))
+ {
+ return false;
+ }
+ obj_value = *obj;
+ return true;
+}
+
+//get value of gradient of objective function at vector x.
+bool minconTMINLP::eval_grad_f(Index n, const Number* x, bool new_x, Number* grad_f)
+{
+ #ifdef DEBUG
+ sciprint("Code is in eval_grad_f\n");
+ #endif
+ char name[20]="_gradf";
+ Number *resg;
+ if (getFunctionFromScilab(n,name,x, 7, 1,2,&resg))
+ {
+ return false;
+ }
+
+ Index i;
+ for(i=0;i<numVars_;i++)
+ {
+ grad_f[i]=resg[i];
+ }
+ return true;
+}
+
+// return the value of the constraints: g(x)
+bool minconTMINLP::eval_g(Index n, const Number* x, bool new_x, Index m, Number* g)
+{
+ #ifdef DEBUG
+ sciprint("Code is in eval_g\n");
+ #endif
+ // return the value of the constraints: g(x)
+ if(m==0)
+ {
+ g=NULL;
+ }
+ else
+ {
+ char name[20]="_addnlc";
+ Number *con;
+ if (getFunctionFromScilab(n,name,x, 7, 1,2,&con))
+ {
+ return false;
+ }
+
+ Index i;
+ for(i=0;i<m;i++)
+ {
+ g[i]=con[i];
+ }
+ }
+
+ return true;
+}
+
+// return the structure or values of the jacobian
+bool minconTMINLP::eval_jac_g(Index n, const Number* x, bool new_x,Index m, Index nele_jac, Index* iRow, Index *jCol,Number* values)
+{
+ #ifdef DEBUG
+ sciprint("Code is in eval_jac_g\n");
+ #endif
+ if (values == NULL)
+ {
+ if(m==0)// return the structure of the jacobian of the constraints
+ {
+ iRow=NULL;
+ jCol=NULL;
+ }
+ else
+ {
+ unsigned int i,j,idx=0;
+ for(i=0;i<m;i++)
+ for(j=0;j<n;j++)
+ {
+ iRow[idx]=i;
+ jCol[idx]=j;
+ idx++;
+ }
+ }
+ }
+ else
+ {
+ if(m==0)
+ {
+ values=NULL;
+ }
+ else
+ {
+ double* resj;
+ char name[20]="_gradnlc";
+ if (getFunctionFromScilab(n,name,x, 7, 1,2,&resj))
+ {
+ return false;
+ }
+ int c = 0;
+ for(int i=0;i<m;i++)
+ {
+ for(int j=0;j<n;j++)
+ {
+ values[c] = resj[j*(int)m+i];
+ c++;
+ }
+ }
+ }
+ }
+ return true;
+}
+
+/*
+ * Return either the sparsity structure of the Hessian of the Lagrangian,
+ * or the values of the Hessian of the Lagrangian for the given values for
+ * x,lambda,obj_factor.
+*/
+
+bool minconTMINLP::eval_h(Index n, const Number* x, bool new_x,Number obj_factor, Index m, const Number* lambda,bool new_lambda, Index nele_hess, Index* iRow,Index* jCol, Number* values)
+{
+ #ifdef DEBUG
+ sciprint("Code is in eval_h\n");
+ #endif
+ double check;
+ if (values==NULL)
+ {
+ Index idx=0;
+ for (Index row = 0; row < numVars_; row++)
+ {
+ for (Index col = 0; col < numVars_; col++)
+ {
+ iRow[idx] = row;
+ jCol[idx] = col;
+ idx++;
+ }
+ }
+ }
+ else
+ { char name[20]="_gradhess";
+ Number *resCh;
+ if (getHessFromScilab(n,m,name,x, &obj_factor, lambda, 7, 3,2,&resCh))
+ {
+ return false;
+ }
+ Index index=0;
+ for (Index row=0;row < numVars_ ;++row)
+ {
+ for (Index col=0; col < numVars_; ++col)
+ {
+ values[index++]=resCh[numVars_*row+col];
+ }
+ }
+ }
+ return true;
+}
+
+void minconTMINLP::finalize_solution(SolverReturn status,Index n, const Number* x, Number obj_value)
+{
+ #ifdef DEBUG
+ sciprint("Code is in finalize_solution\n");
+ sciprint("%d",status);
+ #endif
+ finalObjVal_ = obj_value;
+ status_ = status;
+ if(status==0 ||status== 3)
+ {
+ finalX_ = new double[n];
+ for (Index i=0; i<numVars_; i++)
+ {
+ finalX_[i] = x[i];
+ }
+ }
+}
+
+const double * minconTMINLP::getX()
+{
+ return finalX_;
+}
+
+double minconTMINLP::getObjVal()
+{
+ return finalObjVal_;
+}
+
+int minconTMINLP::returnStatus()
+{
+ return status_;
+}
diff --git a/sci_gateway/cpp/sci_minuncTMINLP.cpp b/sci_gateway/cpp/sci_minuncTMINLP.cpp
new file mode 100644
index 0000000..a3212aa
--- /dev/null
+++ b/sci_gateway/cpp/sci_minuncTMINLP.cpp
@@ -0,0 +1,236 @@
+// Copyright (C) 2015 - IIT Bombay - FOSSEE
+//
+// Author: Harpreet Singh, Pranav Deshpande and Akshay Miterani
+// Organization: FOSSEE, IIT Bombay
+// Email: toolbox@scilab.in
+// This file must be used under the terms of the CeCILL.
+// This source file is licensed as described in the file COPYING, which
+// you should have received as part of this distribution. The terms
+// are also available at
+// http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt
+
+#include "minuncTMINLP.hpp"
+#include "sci_iofunc.hpp"
+
+extern "C"
+{
+#include "call_scilab.h"
+#include <api_scilab.h>
+#include <Scierror.h>
+#include <BOOL.h>
+#include <localization.h>
+#include <sciprint.h>
+#include <string.h>
+#include <assert.h>
+}
+
+using namespace Ipopt;
+using namespace Bonmin;
+
+minuncTMINLP::~minuncTMINLP()
+{
+ if(finalX_) delete[] finalX_;
+}
+
+// Set the type of every variable - CONTINUOUS or INTEGER
+bool minuncTMINLP::get_variables_types(Index n, VariableType* var_types)
+{
+ n = numVars_;
+ for(int i=0; i < n; i++)
+ var_types[i] = CONTINUOUS;
+ for(int i=0 ; i < intconSize_ ; ++i)
+ var_types[(int)(intcon_[i]-1)] = INTEGER;
+ return true;
+}
+
+// The linearity of the variables - LINEAR or NON_LINEAR
+bool minuncTMINLP::get_variables_linearity(Index n, Ipopt::TNLP::LinearityType* var_types)
+{
+ /*
+ n = numVars_;
+ for(int i = 0; i < n; i++)
+ var_types[i] = Ipopt::TNLP::LINEAR;
+ */
+ return true;
+}
+
+// The linearity of the constraints - LINEAR or NON_LINEAR
+bool minuncTMINLP::get_constraints_linearity(Index m, Ipopt::TNLP::LinearityType* const_types)
+{
+ /* m = numConstr_;
+ for(int i = 0; i < m; i++)
+ const_types[i] = Ipopt::TNLP::LINEAR;
+ */
+ return true;
+}
+
+//get NLP info such as number of variables,constraints,no.of elements in jacobian and hessian to allocate memory
+bool minuncTMINLP::get_nlp_info(Index& n, Index& m, Index& nnz_jac_g, Index& nnz_h_lag, TNLP::IndexStyleEnum& index_style)
+{
+ n=numVars_; // Number of variables
+ m=0; // Number of constraints
+ nnz_jac_g = 0; // No. of elements in Jacobian of constraints
+ nnz_h_lag = n*(n+1)/2; // No. of elements in lower traingle of Hessian of the Lagrangian.
+ index_style=TNLP::C_STYLE; // Index style of matrices
+ return true;
+}
+
+//get variable and constraint bound info
+bool minuncTMINLP::get_bounds_info(Index n, Number* x_l, Number* x_u, Index m, Number* g_l, Number* g_u)
+{
+ unsigned int i;
+ for(i=0;i<n;i++)
+ {
+ x_l[i]=-1.0e19;
+ x_u[i]=1.0e19;
+ }
+
+ g_l=NULL;
+ g_u=NULL;
+ return true;
+}
+
+// return the value of the constraints: g(x)
+bool minuncTMINLP::eval_g(Index n, const Number* x, bool new_x, Index m, Number* g)
+{
+ // return the value of the constraints: g(x)
+ g=NULL;
+ return true;
+}
+
+// return the structure or values of the jacobian
+bool minuncTMINLP::eval_jac_g(Index n, const Number* x, bool new_x,Index m, Index nele_jac, Index* iRow, Index *jCol,Number* values)
+{
+ if (values == NULL)
+ {
+ // return the structure of the jacobian of the constraints
+ iRow=NULL;
+ jCol=NULL;
+ }
+ else
+ {
+ values=NULL;
+ }
+
+ return true;
+}
+
+//get value of objective function at vector x
+bool minuncTMINLP::eval_f(Index n, const Number* x, bool new_x, Number& obj_value)
+{
+ char name[20]="_f";
+ Number *obj;
+ if (getFunctionFromScilab(n,name,x, 7, 1,2,&obj))
+ {
+ return false;
+ }
+ obj_value = *obj;
+ return true;
+}
+
+//get value of gradient of objective function at vector x.
+bool minuncTMINLP::eval_grad_f(Index n, const Number* x, bool new_x, Number* grad_f)
+{
+ char name[20]="_gradf";
+ Number *resg;
+ if (getFunctionFromScilab(n,name,x, 7, 1, 2, &resg))
+ {
+ return false;
+ }
+
+ Index i;
+ for(i=0;i<numVars_;i++)
+ {
+ grad_f[i]=resg[i];
+ }
+ return true;
+}
+
+// This method sets initial values for required vectors . For now we are assuming 0 to all values.
+bool minuncTMINLP::get_starting_point(Index n, bool init_x, Number* x,bool init_z, Number* z_L, Number* z_U,Index m, bool init_lambda,Number* lambda)
+{
+ assert(init_x == true);
+ assert(init_z == false);
+ assert(init_lambda == false);
+ if (init_x == true)
+ { //we need to set initial values for vector x
+ for (Index var=0;var<n;var++)
+ x[var]=varGuess_[var];//initialize with 0 or we can change.
+ }
+
+ return true;
+}
+
+/*
+ * Return either the sparsity structure of the Hessian of the Lagrangian,
+ * or the values of the Hessian of the Lagrangian for the given values for
+ * x,lambda,obj_factor.
+*/
+
+bool minuncTMINLP::eval_h(Index n, const Number* x, bool new_x,Number obj_factor, Index m, const Number* lambda,bool new_lambda, Index nele_hess, Index* iRow,Index* jCol, Number* values)
+{
+ double check;
+ if (values==NULL)
+ {
+ Index idx=0;
+ for (Index row = 0; row < numVars_; row++)
+ {
+ for (Index col = 0; col <= row; col++)
+ {
+ iRow[idx] = row;
+ jCol[idx] = col;
+ idx++;
+ }
+ }
+ }
+
+ else
+ {
+ char name[20]="_gradhess";
+ Number *resh;
+ if (getFunctionFromScilab(n,name,x, 7, 1,2,&resh))
+ {
+ return false;
+ }
+ Index index=0;
+ for (Index row=0;row < numVars_ ;++row)
+ {
+ for (Index col=0; col <= row; ++col)
+ {
+ values[index++]=obj_factor*(resh[numVars_*row+col]);
+ }
+ }
+ return true;
+ }
+}
+
+
+void minuncTMINLP::finalize_solution(SolverReturn status,Index n, const Number* x, Number obj_value)
+{
+ finalObjVal_ = obj_value;
+ status_ = status;
+ if(status==0 ||status== 3)
+ {
+ finalX_ = new double[n];
+ for (Index i=0; i<numVars_; i++)
+ {
+ finalX_[i] = x[i];
+ }
+ }
+
+}
+
+const double * minuncTMINLP::getX()
+{
+ return finalX_;
+}
+
+double minuncTMINLP::getObjVal()
+{
+ return finalObjVal_;
+}
+
+int minuncTMINLP::returnStatus()
+{
+ return status_;
+}