I think, I have just made a mistake: I was allocating a static 2D array and accessing it as 1 dimension.
Could you tell me how bad it is - method geta?
The code below works fine on my Windows, and Linux: actual is always eqauls to expected and stride is always equals to N.
#include "stdafx.h"
#define N 2000
int a[N][N];
int geta(int i, int j) {
return *(a[0] + i * N + j);
}
int main()
{
printf("Hello\n");
for (int i = 0; i < N; i++) {
for (int j = 0; j < N; j++) {
a[i][j] = i + j;
}
}
for (int i = 0; i < N; i++) {
for (int j = 0; j < N; j++) {
int const expected = a[i][j];
int const actual = geta(i, j);
if (actual != expected) {
printf("wrong data at [%d,%d] expected=%d actual=%d", i, j, expected, actual);
}
}
}
for (int i = 1; i < N; i++) {
int stride = a[N] - a[N - 1];
if (stride != N) {
printf("wrong: i=%d c=%d N=%d", i, stride, N);
}
}
return 0;
}
Could you tell me how bad it is - method geta?
Bad. But correct. C++ guarantees memory for a to be contiguous and the memory layout to be row major, so your code returns in a correct manner the expected element. Let's see how:
The type of a[0] is int[2000] 1) . But as soon as you do arithmetic on it it decays, i.e. int*. So +i*N moves the pointer to the (beginning of the) line i and +j moves the pointer to the column j.
1) actually it's int(&)[2000] but not that relevant here
Related
So for my homework I have to create a function for adding two matrices, the numbers are given by the user, and I think I have the function somewhat down, I just have no idea what to return because I know you can't return an array, so I tried making a variable to set it equal to but it's not working? Can someone help me out with this?
This is the part I'm stuck on:
int add_matrix(int a[][10], int b[][10], int c[][10], int Anum_rows, int Anum_cols)
{
int matrix_total = c[0][0];
for (int i = 0; i < Anum_rows; ++i) {
for (int j = 0; j < Anum_cols; ++j) {
c[i][j] = a[i][j] + b[i][j];
}
c[i][j] = matrix_total;
}
return matrix_total;
}
The last j in the c array is underlined with red also in visual studio.
Your code already does what it needs, it stores the resulting matrix in c, which, since arrays are passed as pointers, will still contain the updated values when add_matrix() returns. You just have extra code in the function, which should really not be needed to complete the matrix add operation:
void add_matrix(int a[][10], int b[][10], int c[][10], int Anum_rows, int Anum_cols) {
for (int i = 0; i < Anum_rows; ++i) {
for (int j = 0; j < Anum_cols; ++j) {
c[i][j] = a[i][j] + b[i][j];
}
}
}
Good point about Anum_cols. There is no need to pass in the number of columns when it is specified in the argument definition and it could lead to future errors.
#define NUM_COLS 10
void add_matrix(int a[][NUM_COLS], int b[][NUM_COLS], int c[][NUM_COLS], int Anum_rows)
{
for (int i = 0; i < Anum_rows; ++i) {
for (int j = 0; j < NUM_COLS; ++j) {
c[i][j] = a[i][j] + b[i][j];
}
}
}
I have a 2-dimensional array and I am passing it into a function to carry out certain operations. I'd like to know the correct way of doing it...
#define numRows 3
#define numCols 7
#define TotalNum (numRows*numCols)
int arr[numRows][numCols] = {{0,1,2,3,4,5,6}, {7,8,9,10,11,12,13},{14,15,16,17,18,19,20}};
void display(int **p)
{
printf("\n");
for (int i = 0; i< numRows;i++)
{
for ( int j = 0;j< numCols;j++)
{
printf("%i\t",p[i][j]);
}
printf("\n");
}
}
int main() {
display(arr);
}
I get an error message:
'display': cannot convert parameter1 from 'int' to 'int*'
Is this the correct way of passing a 2-dimensional array into a function? If not, what is the correct way?
You should declare your function like this:
void display(int p[][numCols])
This C FAQ thoroughly explains why. The gist of it is that arrays decay into pointers once, it doesn't happen recursively. An array of arrays decays into a pointer to an array, not into a pointer to a pointer.
If (like in your case), you know the dimensions of the array at compilation-time, you can write justvoid display(int p[][numCols]).
Some explanation: You probably know that when you pass an array to a function, you actually pass a pointer to the first member. In C language, 2D array is just an array of arrays. Because of that, you should pass the function a pointer to the first sub-array in the 2D array. So, the natural way, is to say int (*p)[numCols] (that means p is a pointer, to an array of numCols ints). In function declaration, you have the "shortcut" p[], that means exactly the same thing like (*p) (But tells the reader, that you pass a pointer to a beginning of array, and not to just an one variable)
There are several, sometimes equivalent ways of doing this. By declaring an array (cf. method_c()), by using a pointer (cf. method_b()) or by using a pointer to an array of an array (cf. method_a()). method_b(), using a single pointer, is slightly more difficult to get right since it is not easy to use standard array indexing and hence, we use pointer arithmetic. method_a() and method_c() are basically equivalent since arrays decay non-recursively to pointers during compilation. Here is a little program illustrating all three methods. We first initialize a 2x4-array arr in a simple for loop and print it. It will look like this:
arr:
0 1 2 3
0 1 2 3
Afterwards we call all three methods. method_a() adds 1, method_b() adds 2 and method_c() adds 3 to all elements. After each call we print out the array arr again. If a function worked correctly you will easily see it on the output. Size is arbitrary and can be regulated via the two macros ROW and COL. One last note, method_c() relies on variable-length array present since C99.
#include <stdio.h>
#include <stdlib.h>
#define ROW 2
#define COL 4
void method_a(int m, int n, int (*ptr_arr)[n]);
void method_b(int m, int n, int *ptr_arr);
void method_c(int m, int n, int arr[][n]);
int main(int argc, char *argv[]) {
int arr[ROW][COL];
int i;
int j;
for(i = 0; i < ROW; i++) {
for(j = 0; j < COL; j++) {
arr[i][j] = j;
}
}
printf("Original array:\n");
for (i = 0; i < ROW; i++) {
for(j = 0; j < COL; j++) {
printf("%d\t", arr[i][j]);
}
printf("\n");
}
printf("\n\n");
method_a(ROW, COL, arr);
printf("method_a() array:\n");
for (i = 0; i < ROW; i++) {
for(j = 0; j < COL; j++) {
printf("%d\t", arr[i][j]);
}
printf("\n");
}
printf("\n\n");
printf("method_b() array:\n");
method_b(ROW, COL, (int *)arr);
for (i = 0; i < ROW; i++) {
for(j = 0; j < COL; j++) {
printf("%d\t", arr[i][j]);
}
printf("\n");
}
printf("\n\n");
method_c(ROW, COL, arr);
printf("method_c() array:\n");
for (i = 0; i < ROW; i++) {
for(j = 0; j < COL; j++) {
printf("%d\t", arr[i][j]);
}
printf("\n");
}
printf("\n\n");
return EXIT_SUCCESS;
}
void method_a(int m, int n, int (*ptr_arr)[n])
{
int i, j;
for (i = 0; i < m; i++)
{
for (j = 0; j < n; j++)
{
ptr_arr[i][j] = j + 1;
}
}
}
void method_b(int m, int n, int *ptr_arr)
{
int i, j;
for (i = 0; i < m; i++)
{
for (j = 0; j < n; j++)
{
/* We need to use pointer arithmetic when indexing. */
*((ptr_arr + i * n) + j) = j + 2;
}
}
/* The whole function could have also been defined a bit different by taking
* the i index out of the pointer arithmetic. n alone will then provide our
* correct offset to the right. This may be a bit easier to understand. Our
* for-loop would then look like this:
* for (i = 0; i < m; i++)
* {
* for (j = 0; j < n; j++)
* {
* *((ptr_arr + n) + j) = j + 2;
* }
* ptr_arr++;
* }*/
}
void method_c(int m, int n, int arr[][n])
{
int i, j;
for (i = 0; i < m; i++)
{
for (j = 0; j < n; j++)
{
arr[i][j] = j + 3;
}
}
}
You are doing in wrong way. You can pass 2-d array with the help of pointer to an array, or simply pass an array or through Single pointer.
#define numRows 3
#define numCols 7
void display(int (*p)[numcols],int numRows,int numCols)//First method//
void display(int *p,int numRows,int numCols) //Second Method//
void display(int numRows,int numCols,int p[][numCols]) //Third Method
{
printf("\n");
for (int i = 0; i < numRows;i++)
{
for ( int j = 0; j < numCols;j++)
{
printf("%i\t",p[i][j]);
}
printf("\n");
}
}
int main() {
display(arr,numRows,numCols);
}
Declare it simply
void display(int (*p)[numCols][numRows]);
This way your p pointer conveys all necessary informations and you can extract all the dimensions from it without repeating numCols and numRows over and over.
void display(int (*p)[numCols][numRows])
{
size_t i, j;
printf("sizeof array=%zu\n", sizeof *p);
printf("sizeof array[]=%zu\n", sizeof **p);
printf("sizeof array[][]=%zu\n", sizeof ***p);
size_t dim_y = sizeof *p / sizeof **p;
printf("dim_y = %zu\n", dim_y);
size_t dim_x = sizeof **p / sizeof ***p;
printf("dim_x = %zu\n", dim_x);
for(i=0; i<dim_y; i++) {
puts("");
for(j=0; j<dim_x; j++)
printf(" %6d", (*p)[i][j]);
}
}
This is particularly interesting if you use typedefs (which I don't like btw)
typedef int matrix[5][6];
In that case the dimensions are not visible in the signature of the function but the function will still have the correct values for the dimensions.
You can change the signature of display method as follows:
void display(int (*p)[numCols])
Here, p is a pointer to the row of a 2D array. The pointer only needs to know the number of columns in the array.
Actually speaking, pointer need to know the size of each row. This is very important for pointer arithmetic. So that when you increment the pointer, the pointer must point to next row.
Note here, p is not a normal integer pointer. It's a integer pointer to the memory size equal to integer_size x columns.
In main you don't need to change anything. display(arr) is just fine.
Firstly I created my two dimensional array, then I translated it to one dimensional array and I bubble sorted the 1D array, but after I didn't find the pattern to bring it back to 2D array diagonally sorted.
#include<iostream>
#include<iomanip>
const int r = 10;
const int c = 10;
const int lim = r * c;
int A[r][c] = { 0 };
int B[lim];
using namespace std;
void generatearray(int A[][], int r, int c){
srand(time(NULL));
for (int i = 0; i < r; i++)
{
for (int j = 0; j < c; j++)
{
A[i][j] = rand() % lim;
}
}
}
void transformingto1Darray(int A[r][c], int b[lim]){
int p = 0;
for (int m = 0; m < r; m++){
for (int n = 0; n < c; n++){
B[p] = A[m][n];
p++;
}
}
}
void sorting1Darray(int B[][]){
int temp = 0;
for (int k = 0; k < lim - 1; k++){
for (int i = 0; i < lim - 1; i++)
if (B[i] > B[i + 1]){
temp = B[i];
B[i] = B[i + 1];
B[i + 1] = temp;
}
}
}
void sortingdiagonally2Darray(int A[][], int B[]){
int main{
generatearray(A);
transformingto1Darray(A, B);
sorting1Darray(B);
sortingdiagonally2Darray(A, B);
return 0;
}
It's a bit of a wonky solution but it dose work. Because of the way multidimensional indexing works the value in B[i] will be equal to the value in A[0][i].
In your case you want something like this in your sortingdiagonally2Darray function.
for (int i = 0; i > r * c; i++) {
A[0][i] = B[i];
}
This works because under the hood arrays are just pointers. B[x] is syntactic sugar for *(B + x) and A[0][x] will equate to *(*(A + 0) + x) because it's a pointer to a pointer (hence the double star/double brackets).
I have a 2-dimensional array and I am passing it into a function to carry out certain operations. I'd like to know the correct way of doing it...
#define numRows 3
#define numCols 7
#define TotalNum (numRows*numCols)
int arr[numRows][numCols] = {{0,1,2,3,4,5,6}, {7,8,9,10,11,12,13},{14,15,16,17,18,19,20}};
void display(int **p)
{
printf("\n");
for (int i = 0; i< numRows;i++)
{
for ( int j = 0;j< numCols;j++)
{
printf("%i\t",p[i][j]);
}
printf("\n");
}
}
int main() {
display(arr);
}
I get an error message:
'display': cannot convert parameter1 from 'int' to 'int*'
Is this the correct way of passing a 2-dimensional array into a function? If not, what is the correct way?
You should declare your function like this:
void display(int p[][numCols])
This C FAQ thoroughly explains why. The gist of it is that arrays decay into pointers once, it doesn't happen recursively. An array of arrays decays into a pointer to an array, not into a pointer to a pointer.
If (like in your case), you know the dimensions of the array at compilation-time, you can write justvoid display(int p[][numCols]).
Some explanation: You probably know that when you pass an array to a function, you actually pass a pointer to the first member. In C language, 2D array is just an array of arrays. Because of that, you should pass the function a pointer to the first sub-array in the 2D array. So, the natural way, is to say int (*p)[numCols] (that means p is a pointer, to an array of numCols ints). In function declaration, you have the "shortcut" p[], that means exactly the same thing like (*p) (But tells the reader, that you pass a pointer to a beginning of array, and not to just an one variable)
There are several, sometimes equivalent ways of doing this. By declaring an array (cf. method_c()), by using a pointer (cf. method_b()) or by using a pointer to an array of an array (cf. method_a()). method_b(), using a single pointer, is slightly more difficult to get right since it is not easy to use standard array indexing and hence, we use pointer arithmetic. method_a() and method_c() are basically equivalent since arrays decay non-recursively to pointers during compilation. Here is a little program illustrating all three methods. We first initialize a 2x4-array arr in a simple for loop and print it. It will look like this:
arr:
0 1 2 3
0 1 2 3
Afterwards we call all three methods. method_a() adds 1, method_b() adds 2 and method_c() adds 3 to all elements. After each call we print out the array arr again. If a function worked correctly you will easily see it on the output. Size is arbitrary and can be regulated via the two macros ROW and COL. One last note, method_c() relies on variable-length array present since C99.
#include <stdio.h>
#include <stdlib.h>
#define ROW 2
#define COL 4
void method_a(int m, int n, int (*ptr_arr)[n]);
void method_b(int m, int n, int *ptr_arr);
void method_c(int m, int n, int arr[][n]);
int main(int argc, char *argv[]) {
int arr[ROW][COL];
int i;
int j;
for(i = 0; i < ROW; i++) {
for(j = 0; j < COL; j++) {
arr[i][j] = j;
}
}
printf("Original array:\n");
for (i = 0; i < ROW; i++) {
for(j = 0; j < COL; j++) {
printf("%d\t", arr[i][j]);
}
printf("\n");
}
printf("\n\n");
method_a(ROW, COL, arr);
printf("method_a() array:\n");
for (i = 0; i < ROW; i++) {
for(j = 0; j < COL; j++) {
printf("%d\t", arr[i][j]);
}
printf("\n");
}
printf("\n\n");
printf("method_b() array:\n");
method_b(ROW, COL, (int *)arr);
for (i = 0; i < ROW; i++) {
for(j = 0; j < COL; j++) {
printf("%d\t", arr[i][j]);
}
printf("\n");
}
printf("\n\n");
method_c(ROW, COL, arr);
printf("method_c() array:\n");
for (i = 0; i < ROW; i++) {
for(j = 0; j < COL; j++) {
printf("%d\t", arr[i][j]);
}
printf("\n");
}
printf("\n\n");
return EXIT_SUCCESS;
}
void method_a(int m, int n, int (*ptr_arr)[n])
{
int i, j;
for (i = 0; i < m; i++)
{
for (j = 0; j < n; j++)
{
ptr_arr[i][j] = j + 1;
}
}
}
void method_b(int m, int n, int *ptr_arr)
{
int i, j;
for (i = 0; i < m; i++)
{
for (j = 0; j < n; j++)
{
/* We need to use pointer arithmetic when indexing. */
*((ptr_arr + i * n) + j) = j + 2;
}
}
/* The whole function could have also been defined a bit different by taking
* the i index out of the pointer arithmetic. n alone will then provide our
* correct offset to the right. This may be a bit easier to understand. Our
* for-loop would then look like this:
* for (i = 0; i < m; i++)
* {
* for (j = 0; j < n; j++)
* {
* *((ptr_arr + n) + j) = j + 2;
* }
* ptr_arr++;
* }*/
}
void method_c(int m, int n, int arr[][n])
{
int i, j;
for (i = 0; i < m; i++)
{
for (j = 0; j < n; j++)
{
arr[i][j] = j + 3;
}
}
}
You are doing in wrong way. You can pass 2-d array with the help of pointer to an array, or simply pass an array or through Single pointer.
#define numRows 3
#define numCols 7
void display(int (*p)[numcols],int numRows,int numCols)//First method//
void display(int *p,int numRows,int numCols) //Second Method//
void display(int numRows,int numCols,int p[][numCols]) //Third Method
{
printf("\n");
for (int i = 0; i < numRows;i++)
{
for ( int j = 0; j < numCols;j++)
{
printf("%i\t",p[i][j]);
}
printf("\n");
}
}
int main() {
display(arr,numRows,numCols);
}
Declare it simply
void display(int (*p)[numCols][numRows]);
This way your p pointer conveys all necessary informations and you can extract all the dimensions from it without repeating numCols and numRows over and over.
void display(int (*p)[numCols][numRows])
{
size_t i, j;
printf("sizeof array=%zu\n", sizeof *p);
printf("sizeof array[]=%zu\n", sizeof **p);
printf("sizeof array[][]=%zu\n", sizeof ***p);
size_t dim_y = sizeof *p / sizeof **p;
printf("dim_y = %zu\n", dim_y);
size_t dim_x = sizeof **p / sizeof ***p;
printf("dim_x = %zu\n", dim_x);
for(i=0; i<dim_y; i++) {
puts("");
for(j=0; j<dim_x; j++)
printf(" %6d", (*p)[i][j]);
}
}
This is particularly interesting if you use typedefs (which I don't like btw)
typedef int matrix[5][6];
In that case the dimensions are not visible in the signature of the function but the function will still have the correct values for the dimensions.
You can change the signature of display method as follows:
void display(int (*p)[numCols])
Here, p is a pointer to the row of a 2D array. The pointer only needs to know the number of columns in the array.
Actually speaking, pointer need to know the size of each row. This is very important for pointer arithmetic. So that when you increment the pointer, the pointer must point to next row.
Note here, p is not a normal integer pointer. It's a integer pointer to the memory size equal to integer_size x columns.
In main you don't need to change anything. display(arr) is just fine.
Passing matrix as a pointer to pointer to function not working.
#include <stdio.h>
void printMatrix(int **matrix, int row, int col)
{
for (int i = 0; i < row; i++)
{
for (int j = 0; j < col; j++)
printf("%d ", matrix[i][j]);
printf("\r\n");
}
}
void printM (size_t row, size_t col, int matrix[3][4])
{
for (int i = 0; i < row; i++)
{
for (int j = 0; j < col; j++)
printf("%d ", matrix[i][j]);
printf("\r\n");
}
}
int main()
{
int M[3][4];
for (int i = 0; i < 3; i++)
for (int j = 0; j < 4; j++)
M[i][j] = 4*i+j;
printM(3, 4, M);
int *row = *M;
printMatrix(&row, 3, 4); //not working
}
Function printM works, but I would like to know how to use pointer to pointer correctly, thanks for help.
First of all, thank-you for this question. It is a good review of how C does multi-dimensional arrays. Also, it is OK to do double pointers. Remember an array reference is equivalent to a pointer, such as: a[0] and *a both refer to the first element of int a[12]; where *a is the de-referencing of pointer a. And so, &M is a the address of the pointer M when M is declared as int M[3][4];
I modified your code by adding a few comments for clarity and so that it would run in Eclipse using a C compiler from Microsoft, specifically, int declarations where moved out of the for statements. Other than that it is the same as what you originally wrote with a change to the printMatrix declaration and how it is invoked.
Hope this helps, please ask if more questions...
#include <stdio.h>
void printMatrix(int (*matrix)[3][4], int row, int col)
{
int i, j;
// point t so that when de-referenced it is at
// the matrices first element
int *t = (*matrix)[0];
printf("\n");
for (i = 0; i < row; i++)
{
// in C matrices are stored in Row Major form, so
// per K&R just sequentially loop thru (*t)[12]
for (j = 0; j < col; j++) {printf("%d ", *t++);}
printf("\r\n");
}
} // end printMatrix
void printM (size_t row, size_t col, int matrix[3][4])
{
int i, j;
printf("\n");
for (i = 0; i < row; i++)
{
for (j = 0; j < col; j++) {printf("%d ", matrix[i][j]);}
// new line for next row
printf("\r\n");
}
}
int main()
{
int i,j;
// define a matrix with 3 rows and 4 columns
int M[3][4];
// fill-in the matrix with values
for (i = 0; i < 3; i++)
for (j = 0; j < 4; j++)
M[i][j] = 4*i + j;
// print the three rows and four columns of M
printM(3, 4, M);
printMatrix(&M, 3, 4); // Also Works
} // end main
void printMatrix(int *matrix, int row, int col)
{
for (int i = 0; i < row; i++)
{
for (int j = 0; j < col; j++)
printf("%d ", *(matrix+(i*col)+j);
printf("\r\n");
}
}
Don't do a double pointer.