I want to have an variable-length array contained within a structure, but am having trouble initializing it correctly.
struct Grid {
int rows;
int cols;
int grid[];
}
int main() {
struct Grid testgrid = {1, 3, {4, 5, 6}};
}
Everything I try gives me an 'error: non-static initialization of a flexible array member' error.
Here is my version:
#include <stdio.h>
struct matrix {
int rows;
int cols;
int **val;
} a = { .rows=3, .cols=1,
.val = (int*[3]){ (int[1]){1},
(int[1]){2},
(int[1]){3} } },
b = { .rows=3, .cols=4,
.val = (int*[3]){ (int[4]){1, 2, 3, 4},
(int[4]){5, 6, 7, 8},
(int[4]){9,10,11,12} } };
void print_matrix( char *name, struct matrix *m ){
for( int row=0;row<m->rows;row++ )
for( int col=0;col<m->cols;col++ )
printf( "%s[%i][%i]: %i\n", name, row, col, m->val[row][col] );
puts("");
}
int main(){
print_matrix( "a", &a );
print_matrix( "b", &b );
}
You can make that work in gcc by making the struct either static or global, but it turns out that initializing flexible array members is non-conforming and so it is likely to not work except with gcc. Here is a way to do it that just uses C99-conforming features...
#include <stdlib.h>
#include <stdarg.h>
typedef struct Grid {
int rows;
int cols;
int grid[];
} *Grid;
Grid newGrid(int, int, ...);
Grid newGrid(int rows, int cols, ...)
{
Grid g;
va_list ap;
int i, n = rows * cols;
if((g = malloc(sizeof(struct Grid) + rows * cols * sizeof(int))) == NULL)
return NULL;
g->rows = rows;
g->cols = cols;
va_start(ap, cols);
for(i = 0; i < n; ++i)
g->grid[i] = va_arg(ap, int);
va_end(ap);
return g;
}
.
.
.
Grid g1, g2, g3;
g1 = newGrid(1, 1, 123);
g2 = newGrid(2, 3, 1, 1, 1,
2, 2, 2);
g3 = newGrid(4, 5, 1, 2, 3, 4, 5,
6, 7, 8, 9, 10,
11, 12, 13, 14, 15,
16, 17, 18, 19, 20);
You don't have a variable length array (VLA) in your structure. What you have in your structure is called a flexible array member. Flexible array member has absolutely nothing to do with VLA. Flexible array members in C exist to legalize and support the good-old "struct hack" idiom, which is based on dynamic allocation of memory for struct objects with trailing arrays of different size.
Flexible array members cannot be initialized with aggregate initializers, which is what you seem to attempt in your code. What you are trying to do here is simply impossible. There's no such feature in C.
Meanwhile, the text of the error message generated by your compiler seems to suggest that it supports something like this as an extension. This might be true, but keep in mind that this is in no way a standard C feature.
I do not believe that this is possible or supported. As DigitalRoss points out, you can initialize from a literal in the case of static arrays... though I'm still not sure if this is included in the Standard or just a common extension. I can't seem to find a clause in the Standard that supports literal initialization of flexible arrays though I can see that gcc explicitly supports it.
A version using malloc:
#include <stdio.h>
#include <stdlib.h>
typedef struct Grid {
int rows;
int cols;
int *grid;
} Grid;
/* Should validate params */
Grid
buildGrid(int rows, int cols, int vec[]) {
Grid grid;
grid.rows = rows;
grid.cols = cols;
int i;
if ( (grid.grid = malloc(sizeof(vec))) == NULL ) {
/* do something.*/
}
for(i = 0; i < sizeof(vec) ; i++ ) {
grid.grid[i] = vec[i];
}
return grid;
}
Related
I'm trying to assign an array to one of the fields of a typedef struct and I can't find a way to do it practically.
I've searched for this problem but all I seem to find is answers for char * arrays which is not what I'm looking for, I'm just trying to assign an array to an int array, and looking for a practical way for the code below to work without having to initialize all the variables in the struct (they will be initialized later, but I just want to set the array variable):
typedef struct {
int array[5];
int number;
} Rot;
Rot RA;
void config()
{
RA.array = {1, 2, 3, 4, 5}; //This returns an "expected expression before "{"" error
int arr[5];
int i;
for (i = 0; i < 5; i++)
{
arr[i] = i + 1;
}
RA.array = arr; //I understand why this fails, but I need to do this in a practical way
}
Please assume that config is called later and the struct and RA are all accessible to it.
RA.array = {1, 2, 3, 4, 5};
memcpy(RA.array, (int[]){1, 2, 3, 4, 5}, sizeof RA.array);
RA.array = arr;
memcpy(RA.array, arr, sizeof arr); // better: sizeof RA.array
You can use memcpy as shown in another answer. Or alternatively, copy the whole struct and not just the array, using a temporary variable in the form of a compound literal:
RA = (Rot) { {1, 2, 3, 4, 5}, 0 };
This is possible because while C doesn't allow run-time assignment of arrays, it does allow it of structs.
You can use memcpy as shown in another answer, or copy the whole struct as shown in another answer (although your question states that you just want to set the array, not the remainder of the struct).
Another option is to embed just the array into another struct:
typedef struct {
int elem[5];
} RotArr;
typedef struct {
RotArr arr;
int number;
} Rot;
Then you can access element i of the array in Rot RA as RA.arr.elem[i]. Also, you can assign values to a whole RotArr object. The remainder of your code could look something like this:
Rot RA;
void config(void)
{
RA.arr = (RotArr){{1, 2, 3, 4, 5}};
RotArr arr;
int i;
for (i = 0; i < 5; i++)
{
arr.elem[i] = i + 1;
}
RA.arr = arr;
}
Note that (RotArr){{1, 2, 3, 4, 5}} is a compound literal value of RotArr type. It could also be written as (RotArr){ .elem = {1, 2, 3, 4, 5} } or (RotArr){ .elem = { [0] = 1, [1] = 2, [2] = 3, [3] = 4, [4] = 5 } } to be absolutely clear which parts of the compound literal are being set explicitly (any remaining parts will be set to 0), but since it only has a single member, these forms of the compound literal value are a bit over-the-top.
The following works according to C syntax. Not sure this is what you wanted.
#include <string.h>
#include <stdio.h>
typedef struct {
int array[5];
int number;
} Rot;
Rot RA = {{1,2,3,4,5}};
void main()
{
RA = (Rot) {{5, 6, 7, 8, 9}};
int arr[5];
int i;
for (i = 0; i < 5; i++)
{
arr[i] = i + 1;
}
memmove(RA.array, arr, sizeof(RA.array));
// OR
int l = sizeof(arr)/sizeof(arr[0]);
for(int i =0 ; i < l ; ++i) {
*(RA.array + i) = *(arr + i);
printf("%d\n",RA.array[i]);
}
}
Moreover, use memmove since that allows memory overlap.
I'm a novice in C and I need a structure to pass constant two-dimensional arrays to function as one parameter. I want to make this
const int a_size_x = 20;
const int a_size_y = 30;
const int a_output_array[size_x][size_y] = {{..., ...}, ..., {..., ...}};
const int b_size_x = 20;
const int b_size_y = 30;
const int b_output_array[size_x][size_y] = {{..., ...}, ..., {..., ...}};
void function(const int array[], int arr_size_x, int arr_size_y){
for (int i = 0; i < arr_size_x; i++)
{
for (int j = 0; j < arr_size_y; j++)
{
printf("%i ", array[i][j];
}
printf("\n");
}
function(a_output_array, a_size_x, a_size_y);
function(b_output_array, b_size_x, b_size_y);
easier to be able to call function(a) like this:
const struct OUTPUT
{
const int size_x;
const int size_y;
const int array[size_x][size_y];
};
struct OUTPUT a = {.size_x = 20, .size_y = 30, .array = {{...}, ..., {...}};
....
struct OUTPUT z = {.size_x = 30, .size_y = 20, .array = {{...}, ..., {...}};
function(const struct OUTPUT out){
for (int i = 0; i < out.size_x; i++)
{
for (int j = 0; j < out.size_y; j++)
{
printf("%i ", out.array[i][j];
}
printf("\n");
}
function(a);
function(b);
but of course compiler says that size_x and size_y is undeclared in struct declaration.
I've read about flexible array members, but there's dynamic memory allocation needed, and in AVR Harvard architecture malloc can't work in program memory, where i put all this data.
Is there some way to do it in C? May be, in C++?
UPDATE Answer that worked for me - create a one-dimensional array of lenght 2 + width*height where first two members are true width and height and use a pointer to work with this. Here's an example function to print out this array:
char arr [11] =
{
3 // width
3 // height
1, 2, 3,
4, 5, 6,
7, 8, 9
}
void print_array(char *ptr)
{
char width = *ptr++;
char height= *ptr++;
for (int i = 0; i < height; i++)
{
for (int j = 0; j < width; j++)
{
print("%c\t", *ptr++);
}
print("\n");
}
}
print_array(arr);
For most compilers, 2D arrays can be refered to as 1D as such:
matrix[3][3]=[1,2,3
4,5,6
7,8,9]
Index in 1D is calculated by row size*row number. For example: matrix[5]=6.
This means you can pass only 1 parameter, the row length, and by calculating the length of the whole vector you can deduce the 2nd parameter (number of rows).
You can add the row length parameter to the end of your array, and by so passing the array only, if that helps.
When declaring an array with an initializer, the bounds of the array must be constants. A variable with a const qualifier does not qualify as a constant. You can however use a macro which does a text substitution:
#define A_SIZE_X 2
#define A_SIZE_Y 3
const int a_output_array[A_SIZE_X][A_SIZE_Y] = {{3,4,5},{6,7,8}};
#define B_SIZE_X 2
#define B_SIZE_Y 3
const int b_output_array[B_SIZE_X][B_SIZE_Y] = {{1,2,3},{4,5,6}};
When passing a 2D array to a function, the definition must say that it expects a 2D array. Your is expecting const int array[] which is a 1D array.
You can have a function accept arrays with different bounds if the bounds are specified first in the definition:
void function(int arr_size_x, int arr_size_y, const int array[arr_size_x][arr_size_y]) {
You can then call it like this:
function(A_SIZE_X, A_SIZE_Y, a_output_array);
function(B_SIZE_X, B_SIZE_Y, b_output_array);
Side note first, the first snippet has a wrong signature and your compiler should warn you:
void function(const int array[], int arr_size_x, int arr_size_y){
here, array is a pointer to int (in a function signature, an array automatically gets adjusted to a pointer), but for passing a 2d array, you would need a pointer to array of int. Did you test that snippet? I assume it doesn't do what you want.
With C99 and above (assuming the compiler supports VLA, variable length arrays), something like this would be correct:
void function( int arr_size_x, int arr_size_y, const int (*array)[arr_size_y]){
As for your idea with a struct, you could only do it when you keep the second dimension fixed. A C array is contiguous in memory, so to do the indexing correctly, the compiler must know all dimensions except for the first one at compile time. VLAs are an exception to that rule, but you can't declare a VLA statically.
What you can do however is using a flat array and do the 2d indexing yourself, like in this tiny example:
struct outputdata
{
size_t rows;
size_t cols;
int *data;
};
const int a_data[] = {1, 2, 3, 4, 5, 6};
const struct outputdata a = {
.rows = 2,
.cols = 3,
.data = a_data
};
// [...]
void function(const struct outputdata x)
{
for (size_t r = 0; r < x.rows; ++r)
{
for (size_t c = 0; c < x.cols; ++c)
{
printf("%d ", x.data[r*x.cols + c]);
}
}
}
#include <iostream>
#include <math.h>
#include <stdio.h>
#define column 3
#define row 3
#define share 3
int matrix_multiplication(int left_matrix[][column], int right_matrix[][column], int result_matrix[][column], int rows, int cols, int shared);
int A[][3] = {
{1, 2, 3},
{4, 5, 6},
{7, 8, 9}
},
B[][3] = {
{1, 2, 3},
{4, 5, 6},
{7, 8, 9}
}, C[3][3]; //initialize "hard coded" three matrices
int main() {
matrix_multiplication(A, B, C, row, column, share); //passes address of each matrix to function
return 0;
}
int matrix_multiplication(int left_matrix[][column], int right_matrix[][column], int result_matrix[][column], int rows, int cols, int shared) {
int i, j, k;
for (i = 0; i < 3; i++) {
for (j = 0; j < 3; j++) {//stays within first column and row through first iteration
for (k = 0; k < 3; k++)//ensures inner dimensions match with variable k i.e. ixk * kxj =ixj or A*B=C
result_matrix[i][j] += right_matrix[i][k] * left_matrix[k][j]; //C programming has ability to perform matrix multiplication
//Adds result of each operation to C matrix i.e. +=
printf("%d\t", result_matrix[i][j]); //matrix C is composed of rows(i) and columns(j)
}//new tab after each column iteration
printf("\n"); //new line for each row iteration
}
return 0;
}
This code is an good example of passing multidimensional arrays to a function using pointers and printing multidimensional arrays after mutliplication. There are multiple ways to indicate pointers to the compiler. I recommend seeing the "Correct way of passing a 2 dimensional array into a function." for example:
/*void display(int (*p)[numcols],int numRows,int numCols)//First method//
void dispaly(int *p,int numRows,int numCols) //Second Method//
void dispaly (int p[][numCols],int numRows,int numCols) //Third Method*/
Remove the column variable, and add this above the matrix_multiplication function declaration:
#define column 3
(You may also want to rename column to COLUMNS.)
In C++ you can also do this:
static const int column = 3;
or, in C++11:
constexpr int column = 3;
The idea behind all this is that all but the very first size of a multidimensional array must be known at compile time.
To fix the expected primary-expression before ']' token" error, change your inner assignment to something like this:
result_matrix[i][j] += right_matrix[i][k] * left_matrix[k][j];
Also you should initialize result_matrix with 0s first.
Also remove the * from int *result_matrix[][column].
Most modern compilers display warnings if you pass an int instead of an int*. Please enable all warnings in your compiler, recompile, fix them, and update your question stating that the example code compiles cleanly, without warnings.
To print an element of the matrix, you have to specify which element you want to print:
printf("%d\t", result_matrix[i][j]);
I can't believe that your compiler didn't display a warning when you omitted the [i][j]. Warnings are for your benefit: they indicate possible bugs in your code.
Is there any way to pass a multidimensional array to a function without knowing the no of columns.... I mean say I want to print a multidimensional array say a[][9] and b[][3]. If I make a common function say print.
// I have to specify the no of columns right and since
// the no of columns should be same for both actual and
// formal arguments
void print(int a[][])
I have to make different functions for different multidimensional arrays. There should be some way around it.
How to pass unbound multidimensional array?
You have to include all of the array dimensions, except the innermost one (although you probably do want to give the innermost one anyway, so that your function knows when to stop printing). If the dimension is not known at compile time then you can make it a parameter to the function:
void print(size_t m, size_t n, int a[m][n])
{
for ( size_t i = 0; i < m; ++i )
for ( size_t j = 0; j < n; ++j )
printf("%d\n", a[i][j]);
}
Calling the function:
int main(void)
{
int a[][4] = { { 0, 1, 2, 3 }, {8, 7, 6, 5}, {11, 10, 12, 9} };
print(3, 4, a);
return 0;
}
Matt McNabb's answer shows how to use the C99 or C11 variable-length array facilities. There is an alternative that will work with C89 too (which might be a factor if you code on Windows with MSVC), but you still have to tell the function about both dimensions of the array, and you have to do the subscript calculations yourself:
void print(size_t m, size_t n, int *a)
{
size_t i;
size_t j;
for (i = 0; i < m; i++)
{
for (j = 0; j < n; j++)
printf(" %d", a[i * n + j]);
putchar('\n');
}
}
You might call this as:
int main(void)
{
int a[][4] = { { 0, 1, 2, 3 }, {8, 7, 6, 5}, {11, 10, 12, 9} };
print(3, 4, &a[0][0]);
return 0;
}
Sample output:
0 1 2 3
8 7 6 5
11 10 12 9
(One minor comment: I'm not certain that the automatic array could be initialized like that in C89 — there were some restrictions still on automatic variable initialization. If it doesn't work, simply move the entire array declaration outside of main() and prefix it with static so it becomes a file scope array.)
I'm working in C, and am attempting to replicate the length member function that other languages (ie: C++) use to determine the length of an array or possibly vector of data. Is this something I can accomplish in C, or do I have to fall back onto examples like this:
int arraySize = sizeof(array) / sizeof(array[0]);
In general, in C, you use arrays as they are: a contiguous set of multiple pieces of data of the same type. You are generally expected to keep track of array sizes, along with the size of individual members, yourself.
In C++, for example, you have access to the Vector class, which encapsulates and handles all this record keeping for you.
In C, you would be expected to know exactly how big the array is. This is especially important in the case of pointer decay. Your initial example...
int arrayOfInts[6] = {1, 2, 3, 4, 5, 6};
int sizeOfArray = sizeof(arrayOfInts) / sizeof(int); // Success; Returns "6"
This works in this case, but it will fail if you were to pass the array to a function expecting an array of integers (as a pointer) as a function argument.
#include <stdio.h>
int getArraySize(int* arr);
int main(void) {
int arrayOfInts[6] = {1, 2, 3, 4, 5, 6};
int sizeOfArray = getArraySize(arrayOfInts);
return 0;
}
int getArraySize(int* arr) {
int ret;
ret = sizeof(arr) / sizeof(int); // FAILS, returns the size of a pointer-to-int, not the size of the array
return ret;
}
There are two ways to handle this: static definitions, or careful dynamic memory management.
// CASE 1: Trivial case, all arrays are of a static fixed size
#include <stdio.h>
#define ARR_SIZE (6)
int getArraySize(int* arr);
int main(void) {
int arrayOfInts[ARR_SIZE] = {1, 2, 3, 4, 5, 6};
int sizeOfArray = getArraySize(arrayOfInts);
return 0;
}
int getArraySize(int* arr) {
return ret ARR_SIZE;
}
// CASE 2: Managing sizes with dynamic allocation
#include <stdio.h>
#define ARR_SIZE (6)
int main(void) {
int sizeOfArray = ARR_SIZE;
int* arrayOfInts = malloc(sizeOfArray*sizeof(int));
if (arrayOfInts != NULL) {
// Success; initialize
int i;
for (i=0; i<sizeOfArray; i++) {
arrayOfInts[i] = i;
}
return 0;
} else {
// Failed; abort
sizeOfArray = 0;
return (-1);
}
}