org.chocosolver.solver.constraints.nary.lex.PropLexChain Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of choco-solver Show documentation
Show all versions of choco-solver Show documentation
Open-source constraint solver.
/**
* Copyright (c) 2016, Ecole des Mines de Nantes
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the .
* 4. Neither the name of the nor the
* names of its contributors may be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY ''AS IS'' AND ANY
* EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
* WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
* DISCLAIMED. IN NO EVENT SHALL BE LIABLE FOR ANY
* DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
* (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
* ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
* SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
package org.chocosolver.solver.constraints.nary.lex;
import org.chocosolver.solver.constraints.Propagator;
import org.chocosolver.solver.constraints.PropagatorPriority;
import org.chocosolver.solver.exception.ContradictionException;
import org.chocosolver.solver.variables.IntVar;
import org.chocosolver.solver.variables.events.IntEventType;
import org.chocosolver.solver.variables.events.PropagatorEventType;
import org.chocosolver.util.ESat;
import org.chocosolver.util.tools.ArrayUtils;
/**
* Solver constraint of the LexChain constraint.
* Allows to sort lexical chain with strict lexicographic ordering or not.
*
*
* @author Ashish
* @author Charles Prud'homme
* @since 09/08/11
*/
public class PropLexChain extends Propagator {
// the number of variables in each vector of the chain - v1 <= lexChainEq/lexChain <= v2 .....
private int N;
// total number of vector in the lex chain constraint
private int M;
// array for holding lexicographically largest feasible upper bound of each vector
private int[][] UB;
// array for holding lexicographically smallest feasible lower bound of each vector
private int[][] LB;
// If strict's value is true then lexChain is implemented , if false lexChainEq
private boolean strict;
// array of vectors in the lex chain constraint
private IntVar[][] x;
public PropLexChain(IntVar[][] variables, boolean strict) {
super(ArrayUtils.flatten(variables), PropagatorPriority.LINEAR, true);
M = variables.length;
this.N = variables[0].length;
this.x = new IntVar[M][N];
int p = 0;
for (int i = 0; i < M; i++) {
System.arraycopy(vars, p, x[i], 0, N);
p += N;
}
this.strict = strict;
UB = new int[M][N];
LB = new int[M][N];
}
@Override
public int getPropagationConditions(int vIdx) {
return IntEventType.boundAndInst();
}
@Override
public void propagate(int evtmask) throws ContradictionException {
if (PropagatorEventType.isFullPropagation(evtmask)) {
for (int i = 0; i < N; i++) {
UB[M - 1][i] = x[M - 1][i].getUB();
}
for (int i = M - 2; i >= 0; i--) {
computeUB(x[i], UB[i + 1], UB[i]);
}
for (int i = 0; i < N; i++) {
LB[0][i] = x[0][i].getLB();
}
for (int i = 1; i < M; i++) {
computeLB(x[i], LB[i - 1], LB[i]);
}
}
for (int i = 0; i < M; i++) {
boundsLex(LB[i], x[i], UB[i]);
}
}
@Override
public void propagate(int idxVarInProp, int mask) throws ContradictionException {
int vec_idx = idxVarInProp % M;
if (IntEventType.isDecupp(mask)) {
for (int i = 0; i < N; i++) {
UB[vec_idx][i] = x[vec_idx][i].getUB();
}
for (int i = vec_idx - 1; i >= 0; i--) {
computeUB(x[i], UB[i + 1], UB[i]);
}
}
if (IntEventType.isInclow(mask)) {
for (int i = 0; i < N; i++) {
LB[vec_idx][i] = x[vec_idx][i].getLB();
}
for (int i = vec_idx + 1; i < M; i++) {
computeLB(x[i], LB[i - 1], LB[i]);
}
}
forcePropagate(PropagatorEventType.FULL_PROPAGATION);
}
@Override
public ESat isEntailed() {
if (isCompletelyInstantiated()) {
return ESat.eval(checkTuple(0));
}
return ESat.UNDEFINED;
}
/**
* check the feasibility of a tuple, recursively on each pair of consecutive vectors.
* Compare vector xi with vector x(i+1):
* return false if xij > x(i+1)j or if (strict && xi=x(i+1)), and checkTuple(i+1, tuple) otherwise.
*
* @param i the index of the first vector to be considered
* @return true iff lexChain(xi,x(i+1)) && lexChain(x(i+1),..,xk)
*/
private boolean checkTuple(int i) {
if (i == x.length - 1) return true;
int index = N * i;
for (int j = 0; j < N; j++, index++) {
if (vars[index].getValue() > vars[index + N].getValue())
return false;
if (vars[index].getValue() < vars[index + N].getValue())
return checkTuple(i + 1);
}
return (!strict) && checkTuple(i + 1);
}
/////////////////////////////
/**
* Filtering algorithm for between(a,x,b)
* Ensures that x is lexicographically greater than a and less than b if strict is false
* otherwise x is lexicographically greater than or equal to a and less than or equal to b
*
* @param a lexicographically smallest feasible lower bound
* @param x the vector of variables among other vectors in the chain of vectors
* @param b lexicographically largest feasible upper bound
* @throws ContradictionException
*/
private void boundsLex(int[] a, IntVar[] x, int[] b) throws ContradictionException {
int i = 0;
while (i < N && a[i] == b[i]) {
x[i].updateBounds(a[i], b[i], this);
i++;
}
if (i < N) {
x[i].updateBounds(a[i], b[i], this);
}
if (i == N || x[i].nextValue(a[i]) < b[i]) {
return;
}
i++;
while (i < N && x[i].getLB() == b[i] && x[i].getUB() == a[i]) {
if (x[i].hasEnumeratedDomain()) {
x[i].removeInterval(b[i] + 1, a[i] - 1, this);
}
i++;
}
if (i < N) {
if (x[i].hasEnumeratedDomain()) {
x[i].removeInterval(b[i] + 1, a[i] - 1, this);
}
}
}
/**
* computes alpha for use in computing lexicographically largest feasible upper bound of x in
* {@link PropLexChain#computeUB(IntVar[], int[], int[]) computUB}
*
* @param x the vector of variables whose lexicographically largest feasible upper bound is to be computed
* @param b the vector of integers claimed to be the feasible upper bound
* @return an integer greater than or equal to -1 which is used in the computation of lexicographically smallest feasible upper bound vector of integers of x
* @throws ContradictionException
*/
private int computeAlpha(IntVar[] x, int[] b) throws ContradictionException {
int i = 0;
int alpha = -1;
while (i < N && x[i].contains(b[i])) {
if (b[i] > x[i].getLB()) {
alpha = i;
}
i++;
}
if (!strict) {
if (i == N || b[i] > x[i].getLB()) {
alpha = i;
}
} else {
if (i < N && b[i] > x[i].getLB()) {
alpha = i;
}
}
return alpha;
}
/**
* computes beta for use in computing lexicographically smallest feasible lower bound of x in
* {@link PropLexChain#computeLB(IntVar[], int[], int[]) computeLB}
*
* @param x the vector of variables whose lexicographically smallest feasible lower bound is to be computed
* @param a the vector of integers claimed to be the feasible lower bound
* @return an integer greater than or equal to -1 which is used in the computation of lexicographically smallest feasible upper bound vector of integers of x
* @throws ContradictionException
*/
private int computeBeta(IntVar[] x, int[] a) throws ContradictionException {
int i = 0;
int beta = -1;
while (i < N && x[i].contains(a[i])) {
if (a[i] < x[i].getUB()) {
beta = i;
}
i++;
}
if (!strict) {
if (i == N || a[i] < x[i].getUB()) {
beta = i;
}
} else {
if (i < N && a[i] < x[i].getUB()) {
beta = i;
}
}
return beta;
}
/**
* Computes the lexicographically largest feasible upper bound vector of integers of x .
* if aplha computed in {@link PropLexChain#computeAlpha(IntVar[], int[]) computeAlpha} is -1 then
* the current domain values can't satisfy the constraint .So the current intantiations if any are dropped and fresh search is continued.
*
* @param x the vector of variables whose lexicographically largest feasible upper bound is to be computed
* @param b the vector of integers claimed to be the feasible upper bound
* @param u lexicographically largest feasible upper bound of x
* @throws ContradictionException
*/
private void computeUB(IntVar[] x, int[] b, int[] u) throws ContradictionException {
int alpha = computeAlpha(x, b);
if (alpha == -1) fails();
for (int i = 0; i < N; i++) {
if (i < alpha) {
u[i] = b[i];
} else if (i == alpha) {
u[i] = x[i].previousValue(b[i]);
} else {
u[i] = x[i].getUB();
}
}
}
/**
* Computes the lexicographically smallest feasible lower bound vector of integers of x .
* if beta computed in {@link PropLexChain#computeBeta(IntVar[], int[]) computeBeta} is -1 then
* the current domain values can't satisfy the constraint .So the current intantiations if any are dropped and fresh search is continued.
*
* @param x the vector of variables whose lexicographically smallest feasible
* lower bound is to be computed
* @param a the vector of integers claimed to be the feasible lower bound
* @param lower lexicographically smallest feasible lower bound of x
* @throws ContradictionException
*/
private void computeLB(IntVar[] x, int[] a, int[] lower) throws ContradictionException {
int beta = computeBeta(x, a);
if (beta == -1) fails();
for (int i = 0; i < N; i++) {
if (i < beta) {
lower[i] = a[i];
} else if (i == beta) {
lower[i] = x[i].nextValue(a[i]);
} else {
lower[i] = x[i].getLB();
}
}
}
}