import java.util.Arrays;

public class Test {public static final int LEFT_UP = 0;//左上
    public static final int UP_LEFT = 1;//上左
    public static final int UP_RIGHT = 2;//上右
    public static final int RIGHT_UP = 3;//右上
    public static final int RIGHT_DOWN = 4;//右下
    public static final int DOWN_RIGHT = 5;//下右
    public static final int DOWN_LEFT = 6;//下左
    public static final int LEFT_DOWN = 7;//左下
    public static int currentNum = 1;//马的步数本来就位1，因为出生地算一个
    public static int goalNum = 8;//寻找的目标数,如果goalNum为4，则目标数为16

    public static void main(String[] args) {long start = System.currentTimeMillis();
        for (int x = 0; x< goalNum; x++) {for (int y = 0; y< goalNum; y++) {int[] position = {x, y};//马儿的初始位置
                int[][] map = new int[goalNum][goalNum];//地图
                map[position[0]][position[1]] = 1;
                for (int direction = LEFT_UP; direction<= LEFT_DOWN; direction++) {if (move(direction, position, map)) {//如果递归成功
                        long end = System.currentTimeMillis();
                        System.out.println(String.format("总耗时为%d秒", (end - start) / 1000));
                        return;
                    }
                }
            }
        }
    }

    public static boolean judgePosition(int[] position, int[][] map) {//判断路况是否合法即每走完一步，落脚点必须在棋盘内，期盼的边界为   [0,0]-[3,3]，且棋盘当前位置不能重新踩
        return ((position[0] >= 0 && position[0]<= goalNum - 1) && (position[1] >= 0 && position[1]<= goalNum - 1)) && map[position[0]][position[1]] != 1;
    }

    //递归一次，currentNum就得+1
    public static boolean move(int direction, int[] position, int[][] map) {//移动
        if (currentNum == goalNum * goalNum) {//如果当前数量已经达到了递归的目标数，则返回true
            printMap(position, map);
            return true;
        } else {currentNum++;//当前数量+1
            int[] positionCopy = Arrays.copyOf(position, position.length);//一进入这个方法，就将position的值保存下来
            int[][] mapCopy = new int[goalNum][goalNum];
            for (int i = 0; i< goalNum; i++) {for (int j = 0; j< goalNum; j++) {mapCopy[i][j] = map[i][j];
                }
            }
            modifyPosition(direction, position);//修改位置
            if (!judgePosition(position, map)) {//如果修改的坐标不合法或者说原有地图此位置已经为1
                currentNum--;//当前数量-1
                restorePosition(direction, position);//将坐标进行还原
                return false;
            }
            map[position[0]][position[1]] = 1;

            贪心算法,执行速度为0秒//
            int[] sortedDirections = getSortedDirections(position);//经过排序的方向数组
            for (int nextDirection : sortedDirections) {if (move(nextDirection, position, map)) {//如果递归成功
                    printMap(positionCopy, mapCopy);
                    return true;
                }
            }
            

            //无贪心算法/
//            for (int nextDirection = LEFT_UP; nextDirection<= LEFT_DOWN; nextDirection++) {//没贪心算法
//                if (move(nextDirection, position, map)) {//如果递归成功
//                    printMap(positionCopy, mapCopy);
//                    return true;
//                }
//            }
            

            map[position[0]][position[1]] = 0;
            restorePosition(direction, position);//将坐标进行还原
            currentNum--;//当前数量-1
            return false;
        }
    }

    public static int[] getSortedDirections(int[] position) {//将原先代码执行速度从10分钟提升到0秒的关键所在
        int[] directions = {LEFT_UP, UP_LEFT, UP_RIGHT, RIGHT_UP, RIGHT_DOWN, DOWN_RIGHT, DOWN_LEFT, LEFT_DOWN};//八个方向
        for (int i = 0; i< directions.length; i++) {int min = i;//假设当前这个方向是离边界最短距离
            for (int j = i; j< directions.length; j++) {int[] position1 = {position[0], position[1]};
                int positionDistance = 0;
                modifyPosition(directions[j], position1);
                if (position1[0] >3) {positionDistance += goalNum - 1 - position1[0];
                } else {positionDistance += position1[0];
                }
                if (position1[1] >3) {positionDistance += goalNum - 1 - position1[1];
                } else {positionDistance += position1[1];
                }

                int[] position2 = {position[0], position[1]};
                int minPositionDistance = 0;
                modifyPosition(directions[min], position2);
                if (position2[0] >3) {minPositionDistance += goalNum - 1 - position2[0];
                } else {minPositionDistance += position2[0];
                }
                if (position2[1] >3) {minPositionDistance += goalNum - 1 - position2[1];
                } else {minPositionDistance += position2[1];
                }
                if (positionDistance< minPositionDistance) {min = j;
                }
            }
            int temp = directions[i];
            directions[i] = directions[min];
            directions[min] = temp;
        }
        return directions;
    }

    public static void modifyPosition(int direction, int[] position) {//修改位置
        switch (direction) {case LEFT_UP:
                position[0] -= 1;
                position[1] -= 2;
                break;
            case UP_LEFT:
                position[0] -= 2;
                position[1] -= 1;
                break;
            case UP_RIGHT:
                position[0] -= 2;
                position[1] += 1;
                break;
            case RIGHT_UP:
                position[0] -= 1;
                position[1] += 2;
                break;
            case RIGHT_DOWN:
                position[0] += 1;
                position[1] += 2;
                break;
            case DOWN_RIGHT:
                position[0] += 2;
                position[1] += 1;
                break;
            case DOWN_LEFT:
                position[0] += 2;
                position[1] -= 1;
                break;
            case LEFT_DOWN:
                position[0] += 1;
                position[1] -= 2;
                break;
        }
    }

    public static void restorePosition(int direction, int[] position) {//还原位置
        switch (direction) {case LEFT_UP:
                position[0] += 1;
                position[1] += 2;
                break;
            case UP_LEFT:
                position[0] += 2;
                position[1] += 1;
                break;
            case UP_RIGHT:
                position[0] += 2;
                position[1] -= 1;
                break;
            case RIGHT_UP:
                position[0] += 1;
                position[1] -= 2;
                break;
            case RIGHT_DOWN:
                position[0] -= 1;
                position[1] -= 2;
                break;
            case DOWN_RIGHT:
                position[0] -= 2;
                position[1] -= 1;
                break;
            case DOWN_LEFT:
                position[0] -= 2;
                position[1] += 1;
                break;
            case LEFT_DOWN:
                position[0] -= 1;
                position[1] += 2;
                break;
        }
    }

    public static void printMap(int[] position, int[][] map) {for (int i = 0; i< goalNum; i++) {for (int j = 0; j< goalNum; j++) {System.out.print(map[i][j] + "  ");
            }
            System.out.println();
        }
        System.out.println(Arrays.toString(position));
        System.out.println("=======================");
    }
}

你是否还在寻找稳定的海外服务器提供商？创新互联www.cdcxhl.cn海外机房具备T级流量清洗系统配攻击溯源，准确流量调度确保服务器高可用性，企业级服务器适合批量采购，新人活动首月15元起，快前往官网查看详情吧

创新互联公司是一家集网站建设,古冶企业网站建设,古冶品牌网站建设,网站定制,古冶网站建设报价,网络营销,网络优化,古冶网站推广为一体的创新建站企业，帮助传统企业提升企业形象加强企业竞争力。可充分满足这一群体相比中小企业更为丰富、高端、多元的互联网需求。同时我们时刻保持专业、时尚、前沿，时刻以成就客户成长自我，坚持不断学习、思考、沉淀、净化自己，让我们为更多的企业打造出实用型网站。
当前标题：马踏棋盘算法（java+贪心算法）-创新互联
网址分享：http://kswjz.com/article/dcppco.html