受Raymond Chen帖子的启发,假设你有一个4x4二维数组,写一个函数使它旋转90度。Raymond链接到伪代码中的解决方案,但我想看到一些真实的东西。

[1][2][3][4]
[5][6][7][8]
[9][0][1][2]
[3][4][5][6]

就变成:

[3][9][5][1]
[4][0][6][2]
[5][1][7][3]
[6][2][8][4]

更新:Nick的答案是最直接的,但是有没有比n²更好的方法呢?如果矩阵是10000x10000呢?


当前回答

#!/usr/bin/env python

original = [ [1,2,3],
             [4,5,6],
             [7,8,9] ]

# Rotate matrix 90 degrees...
for i in map(None,*original[::-1]):
    print str(i) + '\n'

这导致双方旋转90度(即。123(上面)现在是741(左边)。

这个Python解决方案是可行的,因为它使用了带负步的切片来反转行顺序(将7移到最上面)

original = [ [7,8,9],
             [4,5,6],
             [1,2,3] ]

然后,它使用map(以及隐含的标识函数,这是map以None作为第一个参数的结果)和*按顺序解包所有元素,重新组合列(即。第一个元素一起放在一个元组中,第二个元素一起放在一个元组中,以此类推)。你有效地得到得到返回如下重组:

original = [[7,8,9],
             [4,5,6],
             [1,2,3]]

其他回答

从线性的角度来看,考虑以下矩阵:

    1 2 3        0 0 1
A = 4 5 6    B = 0 1 0
    7 8 9        1 0 0

现在求A

     1 4 7
A' = 2 5 8
     3 6 9

考虑A'对B的作用,或B对A'的作用。 分别为:

      7 4 1          3 6 9
A'B = 8 5 2    BA' = 2 5 8
      9 6 3          1 4 7

这对任何nxn矩阵都是可展开的。 在代码中快速应用这个概念:

void swapInSpace(int** mat, int r1, int c1, int r2, int c2)
{
    mat[r1][c1] ^= mat[r2][c2];
    mat[r2][c2] ^= mat[r1][c1];
    mat[r1][c1] ^= mat[r2][c2];
}

void transpose(int** mat, int size)
{
    for (int i = 0; i < size; i++)
    {
        for (int j = (i + 1); j < size; j++)
        {
            swapInSpace(mat, i, j, j, i);
        }
    }
}

void rotate(int** mat, int size)
{
    //Get transpose
    transpose(mat, size);

    //Swap columns
    for (int i = 0; i < size / 2; i++)
    {
        for (int j = 0; j < size; j++)
        {
            swapInSpace(mat, i, j, size - (i + 1), j);
        }
    }
}

这是c#的

int[,] array = new int[4,4] {
    { 1,2,3,4 },
    { 5,6,7,8 },
    { 9,0,1,2 },
    { 3,4,5,6 }
};

int[,] rotated = RotateMatrix(array, 4);

static int[,] RotateMatrix(int[,] matrix, int n) {
    int[,] ret = new int[n, n];

    for (int i = 0; i < n; ++i) {
        for (int j = 0; j < n; ++j) {
            ret[i, j] = matrix[n - j - 1, i];
        }
    }

    return ret;
}

这是Java中的一个更好的版本:我已经为一个具有不同宽度和高度的矩阵制作了它

H是旋转后矩阵的高度 W是旋转后矩阵的宽度

 

public int[][] rotateMatrixRight(int[][] matrix)
{
    /* W and H are already swapped */
    int w = matrix.length;
    int h = matrix[0].length;
    int[][] ret = new int[h][w];
    for (int i = 0; i < h; ++i) {
        for (int j = 0; j < w; ++j) {
            ret[i][j] = matrix[w - j - 1][i];
        }
    }
    return ret;
}


public int[][] rotateMatrixLeft(int[][] matrix)
{
    /* W and H are already swapped */
    int w = matrix.length;
    int h = matrix[0].length;   
    int[][] ret = new int[h][w];
    for (int i = 0; i < h; ++i) {
        for (int j = 0; j < w; ++j) {
            ret[i][j] = matrix[j][h - i - 1];
        }
    }
    return ret;
}

此代码基于Nick Berardi的帖子。

原地旋转不可能比O(n²)更快,原因是如果我们想旋转矩阵,我们必须至少一次触及所有n²元素,无论你实现什么算法。

虽然旋转数据可能是必要的(也许是为了更新物理存储的表示),但在数组访问上添加一层间接层(也许是一个接口)会变得更简单,可能更性能:

interface IReadableMatrix
{
    int GetValue(int x, int y);
}

如果你的矩阵已经实现了这个接口,那么它可以通过这样一个装饰器类来旋转:

class RotatedMatrix : IReadableMatrix
{
    private readonly IReadableMatrix _baseMatrix;

    public RotatedMatrix(IReadableMatrix baseMatrix)
    {
        _baseMatrix = baseMatrix;
    }

    int GetValue(int x, int y)
    {
        // transpose x and y dimensions
        return _baseMatrix(y, x);
    }
}

旋转+90/-90/180度,水平/垂直翻转和缩放都可以以这种方式实现。

Performance would need to be measured in your specific scenario. However the O(n^2) operation has now been replaced with an O(1) call. It's a virtual method call which is slower than direct array access, so it depends upon how frequently the rotated array is used after rotation. If it's used once, then this approach would definitely win. If it's rotated then used in a long-running system for days, then in-place rotation might perform better. It also depends whether you can accept the up-front cost.

与所有性能问题一样,测量,测量,测量!