返回C中的2D数组的数组(指针)

时间:2021-04-15 21:35:01

A function dynamically creates an int array whose elements are predetermined to be int[2]. Is there any way to have a function assign values to that array and then return it to the caller.

函数动态创建一个int数组,其元素预定为int [2]。有没有办法让函数为该数组赋值,然后将其返回给调用者。

The below code accomplishes this task but throws warning: initialization from incompatible pointer type [enabled by default]

以下代码完成此任务但引发警告:从不兼容的指针类型初始化[默认启用]

#include <stdlib.h>
#include <stdio.h>

int *get_values()
{
    int (*x)[2] = malloc(sizeof(int[2])*3);

    x[0][0] = 1;
    x[0][1] = 2;

    x[1][0] = 11;
    x[1][1] = 12;

    x[2][0] = 21;
    x[2][1] = 22;

    return x;   
}

int main()
{
    int (*x)[2] = get_values();

    int i;
    for (i=0; i!=3; i++)
        printf("x[%d] = { %d, %d }\n", i, x[i][0], x[i][1]);
}

I'm aware of the alternative where you dynamically allocate both dimensions, but this is something that I am curious about.

我知道你动态分配两个维度的替代方案,但这是我很好奇的事情。

2 个解决方案

#1


6  

Rather than keep repeating the same clunky syntax it can be helpful to define a typedef in cases like this. This makes it easier to declare the correct return type for get_values:

而不是继续重复相同的笨重语法,在这种情况下定义typedef会很有帮助。这样可以更容易地为get_values声明正确的返回类型:

#include <stdlib.h>
#include <stdio.h>

typedef int I2[2];

I2 * get_values(void)
{
    I2 * x = malloc(sizeof(I2) * 3);

    x[0][0] = 1;
    x[0][1] = 2;

    x[1][0] = 11;
    x[1][1] = 12;

    x[2][0] = 21;
    x[2][1] = 22;

    return x;   
}

int main()
{
    I2 * x = get_values();

    int i;
    for (i=0; i!=3; i++)
        printf("x[%d] = { %d, %d }\n", i, x[i][0], x[i][1]);

    free(x);
}

LIVE DEMO

现场演示

Recommended reading: Don't repeat yourself (DRY).

推荐阅读:不要重复自己(干)。

#2


3  

And this is how it looks without a typedef:

这就是它没有typedef的样子:

int (*get_values(void))[2]
{  
    return NULL;   
}

Pretty unreadable.

相当难以理解。

Notice in that function definition, if you replace get_values(void) with x you get: int (*x)[2], which is exactly what the pointer definition looks like.

请注意,在该函数定义中,如果用x替换get_values(void),则得到:int(* x)[2],这正是指针定义的样子。

#1


6  

Rather than keep repeating the same clunky syntax it can be helpful to define a typedef in cases like this. This makes it easier to declare the correct return type for get_values:

而不是继续重复相同的笨重语法,在这种情况下定义typedef会很有帮助。这样可以更容易地为get_values声明正确的返回类型:

#include <stdlib.h>
#include <stdio.h>

typedef int I2[2];

I2 * get_values(void)
{
    I2 * x = malloc(sizeof(I2) * 3);

    x[0][0] = 1;
    x[0][1] = 2;

    x[1][0] = 11;
    x[1][1] = 12;

    x[2][0] = 21;
    x[2][1] = 22;

    return x;   
}

int main()
{
    I2 * x = get_values();

    int i;
    for (i=0; i!=3; i++)
        printf("x[%d] = { %d, %d }\n", i, x[i][0], x[i][1]);

    free(x);
}

LIVE DEMO

现场演示

Recommended reading: Don't repeat yourself (DRY).

推荐阅读:不要重复自己(干)。

#2


3  

And this is how it looks without a typedef:

这就是它没有typedef的样子:

int (*get_values(void))[2]
{  
    return NULL;   
}

Pretty unreadable.

相当难以理解。

Notice in that function definition, if you replace get_values(void) with x you get: int (*x)[2], which is exactly what the pointer definition looks like.

请注意,在该函数定义中,如果用x替换get_values(void),则得到:int(* x)[2],这正是指针定义的样子。