/*
* <make_ea_moeo.h>
* Copyright (C) DOLPHIN Project-Team, INRIA Futurs, 2006-2007
* (C) OPAC Team, LIFL, 2002-2007
*
* Arnaud Liefooghe
*
* This software is governed by the CeCILL license under French law and
* abiding by the rules of distribution of free software.  You can  use,
* modify and/ or redistribute the software under the terms of the CeCILL
* license as circulated by CEA, CNRS and INRIA at the following URL
* "http://www.cecill.info".
*
* As a counterpart to the access to the source code and  rights to copy,
* modify and redistribute granted by the license, users are provided only
* with a limited warranty  and the software's author,  the holder of the
* economic rights,  and the successive licensors  have only  limited liability.
*
* In this respect, the user's attention is drawn to the risks associated
* with loading,  using,  modifying and/or developing or reproducing the
* software by the user in light of its specific status of free software,
* that may mean  that it is complicated to manipulate,  and  that  also
* therefore means  that it is reserved for developers  and  experienced
* professionals having in-depth computer knowledge. Users are therefore
* encouraged to load and test the software's suitability as regards their
* requirements in conditions enabling the security of their systems and/or
* data to be ensured and,  more generally, to use and operate it in the
* same conditions as regards security.
* The fact that you are presently reading this means that you have had
* knowledge of the CeCILL license and that you accept its terms.
*
* ParadisEO WebSite : http://paradiseo.gforge.inria.fr
* Contact: paradiseo-help@lists.gforge.inria.fr
*
*/
//-----------------------------------------------------------------------------

#ifndef MAKE_EA_MOEO_H_
#define MAKE_EA_MOEO_H_

#include <stdlib.h>
#include <eoContinue.h>
#include <eoEvalFunc.h>
#include <eoGeneralBreeder.h>
#include <eoGenOp.h>
#include <utils/eoParser.h>
#include <utils/eoState.h>

#include <algo/moeoEA.h>
#include <algo/moeoEasyEA.h>
#include <archive/moeoArchive.h>
#include <comparator/moeoAggregativeComparator.h>
#include <comparator/moeoComparator.h>
#include <comparator/moeoDiversityThenFitnessComparator.h>
#include <comparator/moeoFitnessThenDiversityComparator.h>
#include <diversity/moeoDiversityAssignment.h>
#include <diversity/moeoDummyDiversityAssignment.h>
#include <diversity/moeoFrontByFrontCrowdingDiversityAssignment.h>
#include <diversity/moeoFrontByFrontSharingDiversityAssignment.h>
#include <fitness/moeoDummyFitnessAssignment.h>
#include <fitness/moeoExpBinaryIndicatorBasedFitnessAssignment.h>
#include <fitness/moeoFastNonDominatedSortingFitnessAssignment.h>
#include <fitness/moeoFitnessAssignment.h>
#include <metric/moeoAdditiveEpsilonBinaryMetric.h>
#include <metric/moeoHypervolumeBinaryMetric.h>
#include <metric/moeoNormalizedSolutionVsSolutionBinaryMetric.h>
#include <replacement/moeoElitistReplacement.h>
#include <replacement/moeoEnvironmentalReplacement.h>
#include <replacement/moeoGenerationalReplacement.h>
#include <replacement/moeoReplacement.h>
#include <selection/moeoDetTournamentSelect.h>
#include <selection/moeoRandomSelect.h>
#include <selection/moeoStochTournamentSelect.h>
#include <selection/moeoSelectOne.h>
#include <selection/moeoSelectors.h>


/**
 * This functions allows to build a moeoEA from the parser
 * @param _parser the parser
 * @param _state to store allocated objects
 * @param _eval the funtions evaluator
 * @param _continue the stopping crietria
 * @param _op the variation operators
 * @param _archive the archive of non-dominated solutions
 */
template < class MOEOT >
moeoEA < MOEOT > & do_make_ea_moeo(eoParser & _parser, eoState & _state, eoPopEvalFunc < MOEOT > & _eval, eoContinue < MOEOT > & _continue, eoGenOp < MOEOT > & _op, moeoArchive < MOEOT > & _archive)
{

  /* the objective vector type */
  typedef typename MOEOT::ObjectiveVector ObjectiveVector;


  /* the fitness assignment strategy */
  std::string & fitnessParam = _parser.createParam(std::string("FastNonDominatedSorting"), "fitness",
                               "Fitness assignment scheme: Dummy, FastNonDominatedSorting or IndicatorBased", 'F',
                               "Evolution Engine").value();
  std::string & indicatorParam = _parser.createParam(std::string("Epsilon"), "indicator",
                                 "Binary indicator for IndicatorBased: Epsilon, Hypervolume", 'i',
                                 "Evolution Engine").value();
  double rho = _parser.createParam(1.1, "rho", "reference point for the hypervolume indicator", 'r',
                                   "Evolution Engine").value();
  double kappa = _parser.createParam(0.05, "kappa", "Scaling factor kappa for IndicatorBased", 'k',
                                     "Evolution Engine").value();
  moeoFitnessAssignment < MOEOT > * fitnessAssignment;
  if (fitnessParam == std::string("Dummy"))
    {
      fitnessAssignment = new moeoDummyFitnessAssignment < MOEOT> ();
    }
  else if (fitnessParam == std::string("FastNonDominatedSorting"))
    {
      fitnessAssignment = new moeoFastNonDominatedSortingFitnessAssignment < MOEOT> ();
    }
  else if (fitnessParam == std::string("IndicatorBased"))
    {
      // metric
      moeoNormalizedSolutionVsSolutionBinaryMetric < ObjectiveVector, double > *metric;
      if (indicatorParam == std::string("Epsilon"))
        {
          metric = new moeoAdditiveEpsilonBinaryMetric < ObjectiveVector >;
        }
      else if (indicatorParam == std::string("Hypervolume"))
        {
          metric = new moeoHypervolumeBinaryMetric < ObjectiveVector > (rho);
        }
      else
        {
          std::string stmp = std::string("Invalid binary quality indicator: ") + indicatorParam;
          throw std::runtime_error(stmp.c_str());
        }
      fitnessAssignment = new moeoExpBinaryIndicatorBasedFitnessAssignment < MOEOT > (*metric, kappa);
    }
  else
    {
      std::string stmp = std::string("Invalid fitness assignment strategy: ") + fitnessParam;
      throw std::runtime_error(stmp.c_str());
    }
  _state.storeFunctor(fitnessAssignment);


  /* the diversity assignment strategy */
  eoValueParam<eoParamParamType> & diversityParam = _parser.createParam(eoParamParamType("Dummy"), "diversity",
      "Diversity assignment scheme: Dummy, Sharing(nicheSize) or Crowding", 'D', "Evolution Engine");
  eoParamParamType & diversityParamValue = diversityParam.value();
  moeoDiversityAssignment < MOEOT > * diversityAssignment;
  if (diversityParamValue.first == std::string("Dummy"))
    {
      diversityAssignment = new moeoDummyDiversityAssignment < MOEOT> ();
    }
  else if (diversityParamValue.first == std::string("Sharing"))
    {
      double nicheSize;
      if (!diversityParamValue.second.size())   // no parameter added
        {
          std::cerr << "WARNING, no niche size given for Sharing, using 0.5" << std::endl;
          nicheSize = 0.5;
          diversityParamValue.second.push_back(std::string("0.5"));
        }
      else
        {
          nicheSize = atoi(diversityParamValue.second[0].c_str());
        }
      diversityAssignment = new moeoFrontByFrontSharingDiversityAssignment < MOEOT> (nicheSize);
    }
  else if (diversityParamValue.first == std::string("Crowding"))
    {
      diversityAssignment = new moeoFrontByFrontCrowdingDiversityAssignment < MOEOT> ();
    }
  else
    {
      std::string stmp = std::string("Invalid diversity assignment strategy: ") + diversityParamValue.first;
      throw std::runtime_error(stmp.c_str());
    }
  _state.storeFunctor(diversityAssignment);


  /* the comparator strategy */
  std::string & comparatorParam = _parser.createParam(std::string("FitnessThenDiversity"), "comparator",
                                  "Comparator scheme: FitnessThenDiversity, DiversityThenFitness or Aggregative", 'C', "Evolution Engine").value();
  moeoComparator < MOEOT > * comparator;
  if (comparatorParam == std::string("FitnessThenDiversity"))
    {
      comparator = new moeoFitnessThenDiversityComparator < MOEOT> ();
    }
  else if (comparatorParam == std::string("DiversityThenFitness"))
    {
      comparator = new moeoDiversityThenFitnessComparator < MOEOT> ();
    }
  else if (comparatorParam == std::string("Aggregative"))
    {
      comparator = new moeoAggregativeComparator < MOEOT> ();
    }
  else
    {
      std::string stmp = std::string("Invalid comparator strategy: ") + comparatorParam;
      throw std::runtime_error(stmp.c_str());
    }
  _state.storeFunctor(comparator);


  /* the selection strategy */
  eoValueParam < eoParamParamType > & selectionParam = _parser.createParam(eoParamParamType("DetTour(2)"), "selection",
      "Selection scheme: DetTour(T), StochTour(t) or Random", 'S', "Evolution Engine");
  eoParamParamType & ppSelect = selectionParam.value();
  moeoSelectOne < MOEOT > * select;
  if (ppSelect.first == std::string("DetTour"))
    {
      unsigned int tSize;
      if (!ppSelect.second.size()) // no parameter added
        {
          std::cerr << "WARNING, no parameter passed to DetTour, using 2" << std::endl;
          tSize = 2;
          // put back 2 in parameter for consistency (and status file)
          ppSelect.second.push_back(std::string("2"));
        }
      else // parameter passed by user as DetTour(T)
        {
          tSize = atoi(ppSelect.second[0].c_str());
        }
      select = new moeoDetTournamentSelect < MOEOT > (*comparator, tSize);
    }
  else if (ppSelect.first == std::string("StochTour"))
    {
      double tRate;
      if (!ppSelect.second.size()) // no parameter added
        {
          std::cerr << "WARNING, no parameter passed to StochTour, using 1" << std::endl;
          tRate = 1;
          // put back 1 in parameter for consistency (and status file)
          ppSelect.second.push_back(std::string("1"));
        }
      else // parameter passed by user as StochTour(T)
        {
          tRate = atof(ppSelect.second[0].c_str());
        }
      select = new moeoStochTournamentSelect < MOEOT > (*comparator, tRate);
    }
  /*
  else if (ppSelect.first == string("Roulette"))
  {
      // TO DO !
      // ...
  }
  */
  else if (ppSelect.first == std::string("Random"))
    {
      select = new moeoRandomSelect <MOEOT > ();
    }
  else
    {
      std::string stmp = std::string("Invalid selection strategy: ") + ppSelect.first;
      throw std::runtime_error(stmp.c_str());
    }
  _state.storeFunctor(select);


  /* the replacement strategy */
  std::string & replacementParam = _parser.createParam(std::string("Elitist"), "replacement",
                                   "Replacement scheme: Elitist, Environmental or Generational", 'R', "Evolution Engine").value();
  moeoReplacement < MOEOT > * replace;
  if (replacementParam == std::string("Elitist"))
    {
      replace = new moeoElitistReplacement < MOEOT> (*fitnessAssignment, *diversityAssignment, *comparator);
    }
  else if (replacementParam == std::string("Environmental"))
    {
      replace = new moeoEnvironmentalReplacement < MOEOT> (*fitnessAssignment, *diversityAssignment, *comparator);
    }
  else if (replacementParam == std::string("Generational"))
    {
      replace = new moeoGenerationalReplacement < MOEOT> ();
    }
  else
    {
      std::string stmp = std::string("Invalid replacement strategy: ") + replacementParam;
      throw std::runtime_error(stmp.c_str());
    }
  _state.storeFunctor(replace);


  /* the number of offspring  */
  eoValueParam < eoHowMany > & offspringRateParam = _parser.createParam(eoHowMany(1.0), "nbOffspring",
      "Number of offspring (percentage or absolute)", 'O', "Evolution Engine");


  // the general breeder
  eoGeneralBreeder < MOEOT > * breed = new eoGeneralBreeder < MOEOT > (*select, _op, offspringRateParam.value());
  _state.storeFunctor(breed);
  // the eoEasyEA
  moeoEA < MOEOT > * algo = new moeoEasyEA < MOEOT > (_continue, _eval, *breed, *replace, *fitnessAssignment, *diversityAssignment);
  _state.storeFunctor(algo);
  return *algo;

}

#endif /*MAKE_EA_MOEO_H_*/