/* $Id: CbcHeuristicDivePseudoCost.cpp 1922 2013-05-16 07:35:51Z forrest $ */
// Copyright (C) 2008, International Business Machines
// Corporation and others.  All Rights Reserved.
// This code is licensed under the terms of the Eclipse Public License (EPL).

#if defined(_MSC_VER)
// Turn off compiler warning about long names
#  pragma warning(disable:4786)
#endif

#include "CbcHeuristicDivePseudoCost.hpp"
#include "CbcStrategy.hpp"
#include "CbcBranchDynamic.hpp"

// Default Constructor
CbcHeuristicDivePseudoCost::CbcHeuristicDivePseudoCost()
        : CbcHeuristicDive()
{
}

// Constructor from model
CbcHeuristicDivePseudoCost::CbcHeuristicDivePseudoCost(CbcModel & model)
        : CbcHeuristicDive(model)
{
}

// Destructor
CbcHeuristicDivePseudoCost::~CbcHeuristicDivePseudoCost ()
{
}

// Clone
CbcHeuristicDivePseudoCost *
CbcHeuristicDivePseudoCost::clone() const
{
    return new CbcHeuristicDivePseudoCost(*this);
}

// Create C++ lines to get to current state
void
CbcHeuristicDivePseudoCost::generateCpp( FILE * fp)
{
    CbcHeuristicDivePseudoCost other;
    fprintf(fp, "0#include \"CbcHeuristicDivePseudoCost.hpp\"\n");
    fprintf(fp, "3  CbcHeuristicDivePseudoCost heuristicDivePseudoCost(*cbcModel);\n");
    CbcHeuristic::generateCpp(fp, "heuristicDivePseudoCost");
    fprintf(fp, "3  cbcModel->addHeuristic(&heuristicDivePseudoCost);\n");
}

// Copy constructor
CbcHeuristicDivePseudoCost::CbcHeuristicDivePseudoCost(const CbcHeuristicDivePseudoCost & rhs)
        :
        CbcHeuristicDive(rhs)
{
}

// Assignment operator
CbcHeuristicDivePseudoCost &
CbcHeuristicDivePseudoCost::operator=( const CbcHeuristicDivePseudoCost & rhs)
{
    if (this != &rhs) {
        CbcHeuristicDive::operator=(rhs);
    }
    return *this;
}

bool
CbcHeuristicDivePseudoCost::selectVariableToBranch(OsiSolverInterface* solver,
        const double* newSolution,
        int& bestColumn,
        int& bestRound)
{
    int numberIntegers = model_->numberIntegers();
    const int * integerVariable = model_->integerVariable();
    double integerTolerance = model_->getDblParam(CbcModel::CbcIntegerTolerance);

    // get the LP relaxation solution at the root node
    double * rootNodeLPSol = model_->continuousSolution();

    // get pseudo costs
    double * pseudoCostDown = downArray_;
    double * pseudoCostUp = upArray_;

    bestColumn = -1;
    bestRound = -1; // -1 rounds down, +1 rounds up
    double bestScore = -1.0;
    bool allTriviallyRoundableSoFar = true;
    for (int i = 0; i < numberIntegers; i++) {
        int iColumn = integerVariable[i];
        double rootValue = rootNodeLPSol[iColumn];
        double value = newSolution[iColumn];
        double fraction = value - floor(value);
        int round = 0;
        if (fabs(floor(value + 0.5) - value) > integerTolerance) {
            if (allTriviallyRoundableSoFar || (downLocks_[i] > 0 && upLocks_[i] > 0)) {

                if (allTriviallyRoundableSoFar && downLocks_[i] > 0 && upLocks_[i] > 0) {
                    allTriviallyRoundableSoFar = false;
                    bestScore = -1.0;
                }

                double pCostDown = pseudoCostDown[i];
                double pCostUp = pseudoCostUp[i];
                assert(pCostDown >= 0.0 && pCostUp >= 0.0);

                if (allTriviallyRoundableSoFar && downLocks_[i] == 0 && upLocks_[i] > 0)
                    round = 1;
                else if (allTriviallyRoundableSoFar && downLocks_[i] > 0 && upLocks_[i] == 0)
                    round = -1;
                else if (value - rootValue < -0.4)
                    round = -1;
                else if (value - rootValue > 0.4)
                    round = 1;
                else if (fraction < 0.3)
                    round = -1;
                else if (fraction > 0.7)
                    round = 1;
                else if (pCostDown < pCostUp)
                    round = -1;
                else
                    round = 1;

                // calculate score
                double score;
                if (round == 1)
                    score = fraction * (pCostDown + 1.0) / (pCostUp + 1.0);
                else
                    score = (1.0 - fraction) * (pCostUp + 1.0) / (pCostDown + 1.0);

                // if variable is binary, increase its chance of being selected
                if (solver->isBinary(iColumn))
                    score *= 1000.0;

                if (score > bestScore) {
                    bestColumn = iColumn;
                    bestScore = score;
                    bestRound = round;
                }
            }
        }
    }

    return allTriviallyRoundableSoFar;
}
void
CbcHeuristicDivePseudoCost::initializeData()
{
    int numberIntegers = model_->numberIntegers();
    if (!downArray_) {
        downArray_ = new double [numberIntegers];
        upArray_ = new double [numberIntegers];
    }
    // get pseudo costs
    model_->fillPseudoCosts(downArray_, upArray_);
    // allow for -999 -> force to run
    int diveOptions = (when_>0) ? when_ / 100 : 0;
    if (diveOptions) {
        // pseudo shadow prices
        int k = diveOptions % 100;
        if (diveOptions >= 100)
            k += 32;
        model_->pseudoShadow(k - 1);
        int numberInts = CoinMin(model_->numberObjects(), numberIntegers);
        OsiObject ** objects = model_->objects();
        for (int i = 0; i < numberInts; i++) {
            CbcSimpleIntegerDynamicPseudoCost * obj1 =
                dynamic_cast <CbcSimpleIntegerDynamicPseudoCost *>(objects[i]) ;
            if (obj1) {
                //int iColumn = obj1->columnNumber();
                double downPseudoCost = 1.0e-2 * obj1->downDynamicPseudoCost();
                double downShadow = obj1->downShadowPrice();
                double upPseudoCost = 1.0e-2 * obj1->upDynamicPseudoCost();
                double upShadow = obj1->upShadowPrice();
                downPseudoCost = CoinMax(downPseudoCost, downShadow);
                downPseudoCost = CoinMax(downPseudoCost, 0.001 * upShadow);
                downArray_[i] = downPseudoCost;
                upPseudoCost = CoinMax(upPseudoCost, upShadow);
                upPseudoCost = CoinMax(upPseudoCost, 0.001 * downShadow);
                upArray_[i] = upPseudoCost;
            }
        }
    }
}
// Fix other variables at bounds
int
CbcHeuristicDivePseudoCost::fixOtherVariables(OsiSolverInterface * solver,
        const double * solution,
        PseudoReducedCost * candidate,
        const double * random)
{
    const double * lower = solver->getColLower();
    const double * upper = solver->getColUpper();
    double integerTolerance = model_->getDblParam(CbcModel::CbcIntegerTolerance);
    double primalTolerance;
    solver->getDblParam(OsiPrimalTolerance, primalTolerance);

    int numberIntegers = model_->numberIntegers();
    const int * integerVariable = model_->integerVariable();
    const double* reducedCost = solver->getReducedCost();
    // fix other integer variables that are at their bounds
    int cnt = 0;
    int numberFree = 0;
    int numberFixedAlready = 0;
    for (int i = 0; i < numberIntegers; i++) {
        int iColumn = integerVariable[i];
        if (upper[iColumn] > lower[iColumn]) {
            numberFree++;
            double value = solution[iColumn];
            if (value - lower[iColumn] <= integerTolerance) {
                candidate[cnt].var = iColumn;
                candidate[cnt++].pseudoRedCost = CoinMax(1.0e-2 * reducedCost[iColumn],
                                                 downArray_[i]) * random[i];
            } else if (upper[iColumn] - value <= integerTolerance) {
                candidate[cnt].var = iColumn;
                candidate[cnt++].pseudoRedCost = CoinMax(-1.0e-2 * reducedCost[iColumn],
                                                 downArray_[i]) * random[i];
            }
        } else {
            numberFixedAlready++;
        }
    }
#ifdef CLP_INVESTIGATE
    printf("cutoff %g obj %g - %d free, %d fixed\n",
           model_->getCutoff(), solver->getObjValue(), numberFree,
           numberFixedAlready);
#endif
    return cnt;
    //return CbcHeuristicDive::fixOtherVariables(solver, solution,
    //				     candidate, random);
}