Pass fixed sized array to a function with pointer argument

Question

I'm wondering is there a way to get the following code to work, or do I have to create a new copy of the function for fixed sizes? If so how can I have a generic function for fixed sizes.

void prettyPrintMatrix_float(float **matrix, int rows, int cols){
    int i, j;

    for (i = 0; i<rows; i++){
        for (j = 0; j<cols; j++){
            printf("%10.3f", matrix[i][j]);
        }
        printf("\n");
    }
    return;
}

float ppArray[4][10] = {0.0f};
prettyPrintMatrix_float(ppArray, 4, 10);

Gives an error Access violation reading location 0xFFFFF....

BTW, float[4][10] doesn't decay to float** but (*float)[4]. That is the reason you are getting a violation. — Ajay Brahmakshatriya
– Ajay Brahmakshatriya, Commented Oct 31, 2017 at 14:28
The problem is not passing a fixed-size array to a function, it's passing a two-dimensional array as a pointer to a pointer. float ** is not float[][] — Christian Gibbons
– Christian Gibbons, Commented Oct 31, 2017 at 14:28
@AjayBrahmakshatriya I got that from the warning. But it shouldn't matter in the end because both are saved in the same way in the memory ie as a long 1D array of pointer, right? So is there a way to get around it? — user14492
– user14492, Commented Oct 31, 2017 at 14:31
No, a float ** would be a pointer to a pointer, which could be an array of arrays. A float * could technically point to your matrix, but the issue is that you are now dealing with a decayed pointer; one that does not know how many columns would be in the matrix, so it would not know how to index in two dimension, it would only know how to index in one dimension. — Christian Gibbons
– Christian Gibbons, Commented Oct 31, 2017 at 14:37
@ChristianGibbons What does an array of arrays in look like in the memory? I thought it was just a list of pointer each pointer pointing to each row's first element. — user14492
– user14492, Commented Oct 31, 2017 at 14:45

Vlad from Moscow · Accepted Answer · 2017-10-31 14:44:20Z

2

If the compiler supports Variable Length Arrays then you can define the function like

void prettyPrintMatrix_float( int rows, int cols, float matrix[rows][cols] ){
    int i, j;

    for (i = 0; i<rows; i++){
        for (j = 0; j<cols; j++){
            printf("%10.3f", matrix[i][j]);
        }
        printf("\n");
    }
}

Otherwise you can define the function like

void prettyPrintMatrix_float(const float *matrix, int rows, int cols)
{
    for (int i = 0; i < rows; i++)
    {
        for (int j = 0; j < cols; j++)
        {
            printf("%10.3f", matrix[i * cols + j]);
        }
        putchar('\n');
    }
}

and call it like

prettyPrintMatrix_float( ( const float * )ppArray, 4, 10);

edited Oct 31, 2017 at 14:44

answered Oct 31, 2017 at 14:26

Vlad from Moscow

313k27 gold badges204 silver badges358 bronze badges

Sign up to request clarification or add additional context in comments.

8 Comments

user14492 Over a year ago

I wish. But it does not. So I can't do this. Thanks anyways.

George Over a year ago

@user14492 Huh, are you using a C++ compiler?

George Over a year ago

@user14492 I could be wrong here, but i'm pretty sure VLA's are part of the/mandated by the C99 standard

Christian Gibbons Over a year ago

@George you are correct, C99 standard mandates Variable Length Arrays be supported.

Christian Gibbons Over a year ago

are you passing in the -std=c99 flag?

|

Christian Gibbons · Accepted Answer · 2017-10-31 14:53:52Z

0

If Variable Length Arrays are not available to you, and your function will be taking in matrices of varying dimensions, then you're function will have to treat it as a one-dimensional array and manually do the math for calculating where the start of each row is. Something like this:

void prettyPrintMatrix_float(float *matrix, int rows, int cols){
    int i, j;

    for (i = 0; i<rows; i++){
        for (j = 0; j<cols; j++){
            printf("%10.3f", matrix[i * sizeof(float) + j]);
        }
        printf("\n");
    }
    return;
}

float ppArray[4][10] = {0.0f};
prettyPrintMatrix_float(&ppArray[0][0], 4, 10);

answered Oct 31, 2017 at 14:53

Christian Gibbons

4,3801 gold badge19 silver badges32 bronze badges

Comments

Anton · Accepted Answer · 2023-09-18 15:23:57Z

0

Just another solution with templates

template<int rows, int cols>
void prettyPrintMatrix_float(float (&matrix)[rows][cols])
{
    int i, j;
    for(i = 0; i < rows; i++)
    {
        for(j = 0; j < cols; j++)
        {
            printf("%10.3f", matrix[i,j]);
            // Or
            //printf("%10.3f", matrix[i][j]);
        }
        printf("\n");
    }
    return;
}

answered Sep 18, 2023 at 15:23

Anton

111 bronze badge

1 Comment

Andrew Henle Over a year ago

Just another solution with templates For C?!?!

Collectives™ on Stack Overflow

Pass fixed sized array to a function with pointer argument

3 Answers 3

8 Comments

Comments

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

8 Comments

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related