So. I am working in C and I need some help. I have a matrix(array) (I do not now how to translate it right :D ) that has only 0 and 1 in it. For example, one could look like this:
1 1 0 0 0 0
1 1 0 1 0 0
1 0 0 0 0 1
0 0 1 1 0 1
0 0 1 0 1 1
Now. I need to extract from it the clusters that contain 1. Can you write me some ideas on how to approach this? I tried with a structure and a **pointer to it, with the structure containing 2 elements : x and y, x for x coordinate in the original matrix and y for the y coordinate in the matrix. Then, for each cluster, it would look like :
cluster[0][0].x = 0;
cluster[0][0].y = 0;
cluster[0][1].x = 1;
cluster[0][1].y = 0;
cluster[0][2].x = 0;
cluster[0][2].y = 1;
.. and so on. But I have some problems with the iteration( I have a 1000*1000 matrix) and I decided to ask you if you have any other ideas. Thanks.
EDIT: These are the clusters in this example:
1:
1 1 0 0 0 0
1 1 0 0 0 0
1 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
2:
0 0 0 0 0 0
0 0 0 1 0 0
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
3:
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
0 0 1 1 0 0
0 0 1 0 0 0
4:
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 1
0 0 0 0 0 1
0 0 0 0 1 1
EDIT2:
So. From that matrix of 1 and 0 and I nee to extract all groups of adjacent "1"'s. Adjacent means neighbouring on either up down left or right from it's position. As for the first cluster would be the one made up from those 5 "1"'s from the beginning of the matrix. Another cluster would be that that contains only one "1" on line 2 column 4. And I need somehow to store the x and y coordinate of each cluster somewhere, as I need to use them later on.
For string the data, just an array
char map[1000][1000]
that'll use 1 megabyte of memory, which is not a lot these days.
An algorithm as I see it is
find a 1 in the matrix,
do a flood-fill on it (eg changing 1 to 2 or to 0)
then continue searhcing for a 1 in the matrix.
return the number of fills needed to convert all the 1s.
Flood fill is a well known algorithm, you should be able to find a suitable example, or possibly use a graphics library.
A simple implemention
Use the backtracking to get all clusters, let's start from (0,0) as an example, we first check if (0,0) is 1, if so, check its neighbors one by one. If one of the neighbors is 1, move there and check in the same way. this process doesn't stop until the position's four direction neighbors are all 0 or visited.
To record the position we visited, we need a flag map which has the same size as origin array.
Besides, to draw each cluster, during the backtracking, we record each postion at the same time, I choose a set of list to save all positions in a cluster.
here is all code, including test case you post
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <assert.h>
#define MAX_COL 6
#define MAX_ROW 5
#define MAX_SIZE (MAX_COL * MAX_ROW)
int a[5][6] = {
1, 1, 0, 0, 0, 0,
1, 1, 0, 1, 0, 0,
1, 0, 0, 0, 0, 1,
0, 0, 1, 1, 0, 1,
0, 0, 1, 0, 1, 1,
};
int dir_x[4] = {0, 1, 0, -1};
int dir_y[4] = {1, 0, -1, 0};
struct point {
int x;
int y;
};
struct node {
struct point pos;
struct node *next;
};
struct node* cluster_set[MAX_SIZE];
int cluster_set_index = 0;
int is_inside(int height, int width, int i, int j)
{
if (0 <= j && j < width && i >= 0 && i < height)
return 1;
return 0;
}
int cluster_check(int (*matrix)[MAX_COL], int height, int width, int row, int col, int (*flag_matrix)[MAX_COL], struct node* head)
{
int i, tmp_x, tmp_y;
flag_matrix[row][col] = 1;
for (i = 0; i < 4; i++)
{
tmp_x = row + dir_x[i];
tmp_y = col + dir_y[i];
if (is_inside(height, width, tmp_x, tmp_y) && matrix[tmp_x][tmp_y] && !flag_matrix[tmp_x][tmp_y]) {
flag_matrix[tmp_x][tmp_y] = 1;
struct node *new_node = (struct node*)malloc(sizeof(struct node));
assert(new_node != NULL);
new_node -> pos.x = tmp_x;
new_node -> pos.y = tmp_y;
new_node -> next = NULL;
head -> next = new_node;
cluster_check(matrix, height, width, tmp_x, tmp_y, flag_matrix, new_node);
}
}
}
int cluster_count(int (*matrix)[MAX_COL], int height, int width)
{
int count = 0, i, j;
int flag_matrix[MAX_ROW][MAX_COL] = {0};
for (i = 0; i < height; i++)
for (j = 0; j < width; j++)
{
if (matrix[i][j] && !flag_matrix[i][j]) {
count++;
struct node *new_node = (struct node*)malloc(sizeof(struct node));
assert(new_node != NULL);
new_node -> pos.x = i;
new_node -> pos.y = j;
new_node -> next = NULL;
cluster_set[cluster_set_index++] = new_node;
cluster_check(matrix, height, width, i, j, flag_matrix, new_node);
}
}
return count;
}
void print_cluster(int (*map)[MAX_COL], int row, int col)
{
int i, j;
for (i = 0; i < row; i++)
{
for (j = 0; j < col; j++)
printf("%2d ", map[i][j]);
printf("\n");
}
printf("\n");
}
int main()
{
printf("total clusters: %d\n", cluster_count(a, 5, 6));
int i, cluster_map[MAX_ROW][MAX_COL] = {0};
struct node *tmp;
for (i = 0; i < cluster_set_index; i++)
{
tmp = cluster_set[i];
while (tmp != NULL) {
printf("(%d, %d)", tmp->pos.x, tmp->pos.y);
cluster_map[tmp->pos.x][tmp->pos.y] = 1;
tmp = tmp -> next;
}
printf("\n");
print_cluster(cluster_map, MAX_ROW, MAX_COL);
memset(cluster_map, 0x00, sizeof(int)*MAX_ROW*MAX_COL);
}
}
and here is the running results, just ignore the infomation you don't need
total clusters: 4
(0, 0)(0, 1)(1, 1)(1, 0)(2, 0)
1 1 0 0 0 0
1 1 0 0 0 0
1 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
(1, 3)
0 0 0 0 0 0
0 0 0 1 0 0
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
(2, 5)(3, 5)(4, 5)(4, 4)
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 1
0 0 0 0 0 1
0 0 0 0 1 1
(3, 2)(4, 2)
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
0 0 1 0 0 0
0 0 1 0 0 0
Related
I found here an implementation for Hoshen-Kopelman Algorithm, But it checks neighbors only up and left, meaning that a diagonal connection is not considered a connection.
How can I improve this code so that even a diagonal connection will be considered a connection?
In the following example I expect 1 object and not 7 objects:
4 5
1 0 1 0 1
0 1 0 1 0
1 0 1 0 0
0 0 1 0 0
--input--
1 0 1 0 1
0 1 0 1 0
1 0 1 0 0
0 0 1 0 0
--output--
1 0 2 0 3
0 4 0 5 0
6 0 7 0 0
0 0 7 0 0
HK reports 7 clusters found
This is the implementation (full code can be found here):
#include <stdio.h>
#include <stdlib.h>
#include <assert.h>
/* Implementation of Union-Find Algorithm */
/* The 'labels' array has the meaning that labels[x] is an alias for the label x; by
following this chain until x == labels[x], you can find the canonical name of an
equivalence class. The labels start at one; labels[0] is a special value indicating
the highest label already used. */
int* labels;
int n_labels = 0; /* length of the labels array */
/* uf_find returns the canonical label for the equivalence class containing x */
int uf_find(int x)
{
int y = x;
while (labels[y] != y)
y = labels[y];
while (labels[x] != x)
{
int z = labels[x];
labels[x] = y;
x = z;
}
return y;
}
/* uf_union joins two equivalence classes and returns the canonical label of the resulting class. */
int uf_union(int x, int y)
{
return labels[uf_find(x)] = uf_find(y);
}
/* uf_make_set creates a new equivalence class and returns its label */
int uf_make_set(void)
{
labels[0] ++;
assert(labels[0] < n_labels);
labels[labels[0]] = labels[0];
return labels[0];
}
/* uf_intitialize sets up the data structures needed by the union-find implementation. */
void uf_initialize(int max_labels)
{
n_labels = max_labels;
labels = calloc(sizeof(int), n_labels);
labels[0] = 0;
}
/* uf_done frees the memory used by the union-find data structures */
void uf_done(void)
{
n_labels = 0;
free(labels);
labels = 0;
}
/* End Union-Find implementation */
#define max(a,b) (a>b?a:b)
#define min(a,b) (a>b?b:a)
/* print_matrix prints out a matrix that is set up in the "pointer to pointers" scheme
(aka, an array of arrays); this is incompatible with C's usual representation of 2D
arrays, but allows for 2D arrays with dimensions determined at run-time */
void print_matrix(int** matrix, int m, int n)
{
for (int i = 0; i < m; i++)
{
for (int j = 0; j < n; j++)
printf("%3d ", matrix[i][j]);
printf("\n");
}
}
/* Label the clusters in "matrix". Return the total number of clusters found. */
int hoshen_kopelman(int** matrix, int m, int n)
{
uf_initialize(m * n / 2);
/* scan the matrix */
for (int y = 0; y < m; y++)
{
for (int x = 0; x < n; x++)
{
if (matrix[y][x])
{ // if occupied ...
int up = (y == 0 ? 0 : matrix[y - 1][x]); // look up
int left = (x == 0 ? 0 : matrix[y][x - 1]); // look left
switch (!!up + !!left)
{
case 0:
matrix[y][x] = uf_make_set(); // a new cluster
break;
case 1: // part of an existing cluster
matrix[y][x] = max(up, left); // whichever is nonzero is labelled
break;
case 2: // this site binds two clusters
matrix[y][x] = uf_union(up, left);
break;
}
}
}
}
/* apply the relabeling to the matrix */
/* This is a little bit sneaky.. we create a mapping from the canonical labels
determined by union/find into a new set of canonical labels, which are
guaranteed to be sequential. */
int* new_labels = calloc(sizeof(int), n_labels); // allocate array, initialized to zero
for (int i = 0; i < m; i++)
for (int j = 0; j < n; j++)
if (matrix[i][j])
{
int x = uf_find(matrix[i][j]);
if (new_labels[x] == 0)
{
new_labels[0]++;
new_labels[x] = new_labels[0];
}
matrix[i][j] = new_labels[x];
}
int total_clusters = new_labels[0];
free(new_labels);
uf_done();
return total_clusters;
}
/* This procedure checks to see that any occupied neighbors of an occupied site
have the same label. */
void check_labelling(int** matrix, int m, int n)
{
int N, S, E, W;
for (int i = 0; i < m; i++)
for (int j = 0; j < n; j++)
if (matrix[i][j])
{
N = (i == 0 ? 0 : matrix[i - 1][j]);
S = (i == m - 1 ? 0 : matrix[i + 1][j]);
E = (j == n - 1 ? 0 : matrix[i][j + 1]);
W = (j == 0 ? 0 : matrix[i][j - 1]);
assert(N == 0 || matrix[i][j] == N);
assert(S == 0 || matrix[i][j] == S);
assert(E == 0 || matrix[i][j] == E);
assert(W == 0 || matrix[i][j] == W);
}
}
/* The sample program reads in a matrix from standard input, runs the HK algorithm on
it, and prints out the results. The form of the input is two integers giving the
dimensions of the matrix, followed by the matrix elements (with data separated by
whitespace).
a sample input file is the following:
8 8
1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 1
1 0 0 0 0 1 0 1
1 0 0 1 0 1 0 1
1 0 0 1 0 1 0 1
1 0 0 1 1 1 0 1
1 1 1 1 0 0 0 1
0 0 0 1 1 1 0 1
this sample input gives the following output:
--input--
1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 1
1 0 0 0 0 1 0 1
1 0 0 1 0 1 0 1
1 0 0 1 0 1 0 1
1 0 0 1 1 1 0 1
1 1 1 1 0 0 0 1
0 0 0 1 1 1 0 1
--output--
1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 1
2 0 0 0 0 2 0 1
2 0 0 2 0 2 0 1
2 0 0 2 0 2 0 1
2 0 0 2 2 2 0 1
2 2 2 2 0 0 0 1
0 0 0 2 2 2 0 1
HK reports 2 clusters found
*/
int main(int argc, char** argv)
{
int m, n;
int** matrix;
/* Read in the matrix from standard input
The whitespace-deliminated matrix input is preceeded
by the number of rows and number of columns */
while (2 == scanf_s("%d %d", &m, &n))
{ // m = rows, n = columns
matrix = (int**)calloc(m, sizeof(int*));
for (int i = 0; i < m; i++)
{
matrix[i] = (int*)calloc(n, sizeof(int));
for (int j = 0; j < n; j++)
scanf_s("%d", &(matrix[i][j]));
}
printf_s(" --input-- \n");
print_matrix(matrix, m, n);
printf(" --output-- \n");
/* Process the matrix */
int clusters = hoshen_kopelman(matrix, m, n);
/* Output the result */
print_matrix(matrix, m, n);
check_labelling(matrix, m, n);
printf("HK reports %d clusters found\n", clusters);
for (int i = 0; i < m; i++)
free(matrix[i]);
free(matrix);
}
return 0;
}
I tried to change the function hoshen_kopelman as described below, but I still get 2 objects instead of 1:
int hoshen_kopelman(int** matrix, int m, int n)
{
uf_initialize(m * n / 2);
/* scan the matrix */
for (int y = 0; y < m; y++)
{
for (int x = 0; x < n; x++)
{
if (matrix[y][x])
{ // if occupied ...
int up = (y == 0 ? 0 : matrix[y - 1][x]); // look up
int left = (x == 0 ? 0 : matrix[y][x - 1]); // look left
// ----------- THE NEW CODE -------------
if (x > 0)
{
if (up == 0 && y > 0) // left+up
up = matrix[y - 1][x - 1];
if (left == 0 && y < m - 1) // left+down
left = matrix[y + 1][x - 1];
}
// ---------- END NEW CODE --------------
switch (!!up + !!left)
{
case 0:
matrix[y][x] = uf_make_set(); // a new cluster
break;
case 1: // part of an existing cluster
matrix[y][x] = max(up, left); // whichever is nonzero is labelled
break;
case 2: // this site binds two clusters
matrix[y][x] = uf_union(up, left);
break;
}
}
}
}
/* apply the relabeling to the matrix */
/* This is a little bit sneaky.. we create a mapping from the canonical labels
determined by union/find into a new set of canonical labels, which are
guaranteed to be sequential. */
int* new_labels = calloc(sizeof(int), n_labels); // allocate array, initialized to zero
for (int i = 0; i < m; i++)
for (int j = 0; j < n; j++)
if (matrix[i][j])
{
int x = uf_find(matrix[i][j]);
if (new_labels[x] == 0)
{
new_labels[0]++;
new_labels[x] = new_labels[0];
}
matrix[i][j] = new_labels[x];
}
int total_clusters = new_labels[0];
free(new_labels);
uf_done();
return total_clusters;
}
The following output is now obtained (I am expecting 1 and got 2):
4 5
1 0 1 0 1
0 1 0 1 0
1 0 1 0 0
0 0 1 0 0
--input--
1 0 1 0 1
0 1 0 1 0
1 0 1 0 0
0 0 1 0 0
--output--
1 0 1 0 1
0 1 0 1 0
2 0 1 0 0
0 0 1 0 0
HK reports 2 clusters found
What is the correct way to correct the code to check all 8 neighbors?
I led you astray saying to check down-left. The algorithm relies on the current node it is examining being after all the neighbors it checks. So you need to check left, up, up-left, and up-right. You can use this in place of your new code:
if (y > 0)
{
if (left == 0 && x > 0) // left+up
left = matrix[y - 1][x - 1];
if (up == 0 && x < n-1) // right+up
up = matrix[y - 1][x + 1];
}
My input looks like this :
15 5
0 0 1 0 0
0 0 1 0 0
0 0 1 0 0
0 0 0 0 0
0 0 0 0 0
1 0 0 0 0
0 1 0 1 0
0 0 1 0 0
0 0 0 0 1
0 0 0 0 0
0 0 0 0 0
0 1 0 0 0
1 0 0 0 1
0 0 1 0 0
0 0 0 1 0
The first row contains the number of rows and columns of my array. And basically I want to find out where these 1s are in the array.
So in the first 3 rows i want to get 3, in the 7th 1, and in the 8th, i want to get 2 3 etc..
My code looks like this so far
#include <stdio.h>
int main() {
int row, column;
FILE* input;
FILE* output;
input = fopen("input.txt", "r");
if (input == 0) {
printf("ERROR couldn't open input.txt");
return 1;
}
if (! fscanf(input, "%d %d", &row, &column)) {
printf("ERROR not recognised value");
fclose(input);
return 2;
}
output = fopen("output.txt", "w");
int meteor[row][column];
for (int i = 0; i < row; i++) {
for (int j = 0; j < column; j++) {
fscanf(input, "%d", &meteor[i][j]);
}
}
int sum;
int loc[row];
for (int i = 0; i < row; i++) {
sum = 0;
for (int j = 0; j < column; j++) {
sum += meteor[i][j];
if (meteor[i][j] == 1) {
loc[i] = (j + 1);
}
}
printf("%d %d\n", loc[i], sum);
}
fclose(input);
fclose(output);
return 0;
}
My output is this :
3 1
3 1
3 1
0 0
-1 0
1 1
4 2
3 1
5 1
0 0
4214921 0
2 1
5 2
3 1
4 1
The first column shows some of the locations, the second shows how many 1s are in the row, but it all fails when there are only 0s in the row or there is more than one 1. And also I would like to store these values.
You need to initialize loc[].. If you look carefully at your code, you only populate it on the condition of if (meteor[i][j] == 1)... but you print it for every index of i. which would print uninitialized memory (i.e. unknown).
To answer the second part of your question, if you want to store the "sum". Simply make loc[] a 2d array just like meteor, but with 2 columns (row and sum). i.e. int loc[row][2].. making sure to initialize both columns of course :)
I have a 2D array storing image data using int. At this time it is 800x640 but that can change. I want to pass it to another function in 8x8 blocks for processing. I could actually just copy an 8x8 block of the array into a temporary variable and send that to the function and then copy result into another 800x640 array.
However, I want to the function to directly be able to access 8x8 blocks (which will be faster) if I give it the start xy coordinates within this 800x640 array. The problem is that using int** does not work. Also parameter declared as int[8][8] also does not compiled. What do I do? Right now I am writing the program in C++ but eventually shall have to write it in C as well.
You can give the pointer to the original image with other parameters to the function and access each element of your 8x8 area inside the function.
Let's say this is your original 800x640 image:
int img[640][800];
Declare your fuction as:
void work_on_roi(int* img, size_t img_width, size_t img_height, int roi_x, int roi_y, size_t roi_width, size_t roi_height)
ROI stands for region of interest, a widely used term in the field of image processing. In your case, if you want to access roi with (10,20) as its top-left index, you can call this function with arguments as:
work_on_roi(img, 800, 640, 10, 20, 8, 8)
Inside this function, accessing the (i,j) element in the roi would be:
(img + (roi_y + j) * img_width)[roi_x + i]
You can utilize roi_width and roi_height parameter to check for integrity:
// before accessing (i,j) element of roi
assert(i < roi_width);
assert(j < roi_height);
assert(roi_x + i < img_width);
assert(roi_y + j < img_height);
While the way you access the elements of the region will not change depending on how you have declared your array, the way you pass the array as a parameter will change depending on whether you have an actual 2D array or whether you have a pointer-to-pointer-to-type.
In the case of a true array declared similar to int array[X][Y]; (where X and Y are defined constants) you can pass a pointer to array of int [Y], (e.g. (*array)[Y]) as the parameter to your function.
In the case where array is will be converted to a pointer-to-pointer-to-type, when declared similar to int **array; or int (*array)[z]; where you allocate pointers and blocks of each row, or one single block, respectively, you simply pass a pointer-to-pointer-to-type (e.g. int **array)
Taking either case, you could change a region within the array with a simple function that iterates over the elements you wish to change. For example for the case where you have a 2D array as you specify, you could declare a function with logic similar to the following. (You could pass additional parameters as needed to effect whatever change you need)
enum { ROW = 10, COL = 10 }; /* constant definitions */
...
void chgregion (int (*a)[COL], int xs, int ys, int xn, int yn)
{
int xlim = xs + xn, /* xstart + xnumber_of_elements */
ylim = ys + yn; /* same for y */
if (xlim > ROW) xlim = ROW; /* protect array/block bounds */
if (ylim > COL) ylim = COL;
for (int i = xs; i < xlim; i++)
for (int j = ys; j < ylim; j++)
a[i][j] = 1; /* change element as required */
}
Above the a pointer to an array of COL elements is passed along with the x and y starting position within the array and the number of elements in the region, e.g. xn and yn. A simple check is done to limit the region size to remain within the array bounds or bounds of a block of memory. If your array is actually a pointer-to-pointer-to-type, just pass int **a instead and pass the dimensions of the block of memory as additional parameters.
You can put together a simple test as follows:
#include <stdio.h>
enum { ROW = 10, COL = 10 };
void chgregion (int (*a)[COL], int xs, int ys, int xn, int yn);
void prna (int (*a)[COL]);
int main (void) {
int a[ROW][COL] = {{0}};
prna (a);
chgregion (a, 2, 2, 6, 6);
putchar ('\n');
prna (a);
return 0;
}
void chgregion (int (*a)[COL], int xs, int ys, int xn, int yn)
{
int xlim = xs + xn,
ylim = ys + yn;
if (xlim > ROW) xlim = ROW;
if (ylim > COL) ylim = COL;
for (int i = xs; i < xlim; i++)
for (int j = ys; j < ylim; j++)
a[i][j] = 1;
}
void prna (int (*a)[COL])
{
for (int i = 0; i < ROW; i++) {
for (int j = 0; j < COL; j++)
printf ("%2d", a[i][j]);
putchar ('\n');
}
}
Example Use/Output
$ ./bin/array2d_region
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 1 1 1 1 1 1 0 0
0 0 1 1 1 1 1 1 0 0
0 0 1 1 1 1 1 1 0 0
0 0 1 1 1 1 1 1 0 0
0 0 1 1 1 1 1 1 0 0
0 0 1 1 1 1 1 1 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
Let me know if this is what you were intending, or if what you are doing differs in some way. For further help, please post a Minimal, Complete, and Verifiable example.
I've studied data structure and algorithm, and I got a problem with dynamic 2d array. Here is part of my code. The coding problem is the Knight's tour.
int iMove[8] = {-2, -1, 1, 2, 2, 1, -1, -2};
int jMove[8] = { 1, 2, 2, 1,-1, -2, -2, -1};
cell* cellList(int* i, int* j, int** board){
int k;
cell* temp;
int iTempNext; int jTempNext;
int maxSampleNum = 8;
int cnt = 0;
int val;
for(k = 0; k < maxSampleNum; k++){
iTempNext = (*i) + iMove[k];
jTempNext = (*j) + jMove[k];
//1. get list 0<=i<=7 && 0<=j<=7
if( (0 <= iTempNext && iTempNext <= 7) && (0 <= jTempNext && jTempNext <= 7)){
//2. get the 0 value cells
//val = canMove(iTempNext, jTempNext, board);
printf("%d %d\n", iTempNext, jTempNext);
if(board[iTempNext][jTempNext] == 0){
cell tempCell;
tempCell.row = iTempNext;
tempCell.col = jTempNext;
temp = (cell*)realloc(temp, sizeof(cell));
*(temp+cnt) = tempCell;
cnt++;
}
}
}
return temp;
}
int** board : 2d array and I allocated it dynamically and initialized all elements of this array to 0. I printed 2d array after initializing.
The problem is that after 2nd for loop, this program got segmentation fault on accessing 2d array. I cannot access 2d array elements during 3rd for loop.
Initializing Complete...
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
Enter the start position (i, j): 3 3
iTempNext jTempNext
1 4
2 5
4 5
Program received signal SIGSEGV, Segmentation fault.
0x0000000000400b1c in cellList (i=0x7fffffffe5bc, j=0x7fffffffe5c0, board=0x603010) at main.c:144
144 if(board[iTempNext][jTempNext] == 0){
this is the result after using gdb. How can I solve this problem
This line:
temp = (cell*)realloc(temp, sizeof(cell));
invokes undefined behaviour since temp is not initialized. You should initialize it with, e.g., cell *temp = NULL; (understand that realloc can either take a NULL pointer in which case it is equivalent to malloc, or a previously malloced/calloced/realloced pointer).
And don't cast the return from malloc. This is C, not C++. Search this site for why casting the malloc return value is frowned upon in C.
When main runs, I've been getting this output. The first columns being somehow confused with the last one, even though it's specifically initialized.
0 0 0 0 0 0 4 3 13 3 3 0 0 0 0 0 0
0 0 0 0 0 4 0 0 5 0 0 2 0 0 0 0 0
0 0 0 0 4 0 0 0 5 0 0 0 2 0 0 0 0
0 0 0 18 0 0 0 0 5 0 0 0 0 17 0 0 0
0 0 4 0 2 0 7 12 19 12 3 0 8 0 2 0 0
0 4 0 0 0 0 0 0 5 0 0 0 0 0 0 2 0
0 0 0 0 5 0 0 0 5 0 0 0 5 0 0 0 2
2 0 0 0 11 0 0 0 5 0 0 0 11 0 0 0 1
1 7 7 7 20 7 7 7 14 7 7 7 20 7 7 7 1
1 0 0 0 11 0 0 0 5 0 0 0 11 0 0 0 1
1 0 0 0 1 0 0 0 5 0 0 0 1 0 0 0 0
0 6 0 0 0 0 0 0 5 0 0 0 0 0 0 8 0
0 0 6 0 4 0 7 12 19 12 3 0 6 0 8 0 0
0 0 0 15 0 0 0 0 5 0 0 0 0 16 0 0 0
0 0 0 0 6 0 0 0 5 0 0 0 8 0 0 0 0
0 0 0 0 0 6 0 0 5 0 0 8 0 0 0 0 0
0 0 0 0 0 0 7 7 7 7 8 0 0 0 0 0 0
motion.h
struct square{
int directions;
int isRobotHere;
int isMultipleDirections;
int printable;
};
typedef struct square Square;
struct robot{
int robotx;
int roboty;
int robotz;
int destinationx;
int destinationy;
int destinationz;
};
typedef struct robot Robot;
enum direction{North, NorthWest, West, SouthWest, South, SouthEast, East, NorthEast};
Square firstfloor[16][16];
Square secondfloor[16][16];
void printbothfloors();
void initializeArrays();
initializeArrays.c
#include <stdio.h>
#include "motion.h"
void initializeArrays(){
firstfloor[6][0].directions=5;
firstfloor[7][0].directions=5;
firstfloor[8][0].directions=14;
firstfloor[9][0].directions=5;
firstfloor[10][0].directions=6;
firstfloor[11][1].directions=6;
firstfloor[12][2].directions=6;
firstfloor[13][3].directions=15;
firstfloor[14][4].directions=6;
firstfloor[15][5].directions=6;
firstfloor[16][6].directions=7;
firstfloor[16][7].directions=7;
firstfloor[16][8].directions=7;
firstfloor[16][9].directions=7;
firstfloor[16][10].directions=8;
firstfloor[15][11].directions=8;
firstfloor[14][12].directions=8;
firstfloor[13][13].directions=16;
firstfloor[12][14].directions=8;
firstfloor[11][15].directions=8;
firstfloor[10][16].directions=1;
firstfloor[9][16].directions=1;
firstfloor[8][16].directions=1;
firstfloor[7][16].directions=1;
firstfloor[6][16].directions=2;
firstfloor[5][15].directions=2;
firstfloor[4][14].directions=2;
firstfloor[3][13].directions=17;
firstfloor[2][12].directions=2;
firstfloor[1][11].directions=2;
firstfloor[0][10].directions=3;
firstfloor[0][9].directions=3;
firstfloor[0][8].directions=13;
firstfloor[0][7].directions=3;
firstfloor[0][6].directions=4;
firstfloor[1][5].directions=4;
firstfloor[2][4].directions=4;
firstfloor[3][3].directions=18;
firstfloor[4][2].directions=4;
firstfloor[5][1].directions=4;
firstfloor[1][8].directions=5;
firstfloor[2][8].directions=5;
firstfloor[3][8].directions=5;
firstfloor[4][8].directions=19;
firstfloor[5][8].directions=5;
firstfloor[6][8].directions=5;
firstfloor[7][8].directions=5;
firstfloor[8][8].directions=14;
firstfloor[9][8].directions=5;
firstfloor[10][8].directions=5;
firstfloor[11][8].directions=5;
firstfloor[12][8].directions=19;
firstfloor[13][8].directions=5;
firstfloor[14][8].directions=5;
firstfloor[15][8].directions=5;
firstfloor[8][1].directions=7;
firstfloor[8][2].directions=7;
firstfloor[8][3].directions=7;
firstfloor[8][4].directions=20;
firstfloor[8][5].directions=7;
firstfloor[8][6].directions=7;
firstfloor[8][7].directions=7;
firstfloor[8][9].directions=7;
firstfloor[8][10].directions=7;
firstfloor[8][11].directions=7;
firstfloor[8][12].directions=20;
firstfloor[8][13].directions=7;
firstfloor[8][14].directions=7;
firstfloor[8][15].directions=7;
firstfloor[7][4].directions=11;
firstfloor[9][4].directions=11;
firstfloor[4][7].directions=12;
firstfloor[4][9].directions=12;
firstfloor[12][7].directions=12;
firstfloor[12][9].directions=12;
firstfloor[7][12].directions=11;
firstfloor[9][12].directions=11;
firstfloor[4][4].directions=2;
firstfloor[4][6].directions=7;
firstfloor[4][10].directions=3;
firstfloor[4][12].directions=8;
firstfloor[6][4].directions=5;
firstfloor[6][12].directions=5;
firstfloor[10][4].directions=1;
firstfloor[10][12].directions=1;
firstfloor[12][4].directions=4;
firstfloor[12][6].directions=7;
firstfloor[12][10].directions=3;
firstfloor[12][12].directions=6;
firstfloor[11][0].directions=0;
firstfloor[5][16].directions=0;
}
printbothfloors.c
#include <stdio.h>
#include <stdlib.h>
#include "motion.h"
void printbothfloors(){
// printf("printfloor is running");
int upper, lower, i, j;
printf("%4d %4d %4d %4d %4d", firstfloor[6][0].directions, firstfloor[7][0].directions, firstfloor[8][0].directions, firstfloor[9][0].directions, firstfloor[10][0].directions);
printf("%4d %4d %4d %4d %4d", firstfloor[6][16].directions, firstfloor[7][16].directions, firstfloor[8][16].directions, firstfloor[9][16].directions, firstfloor[10][16].directions);
printf("FIRST FLOOR");
printf("\n-");
/* The next for loop prints out the upper edge */
for (upper = 0; upper < 18; upper++){
printf("----");
}
printf("\n");
/*The next for loop prints out the floor, every element is 4 digits wide */
for (i = 0; i <= 16; i++){
printf("|");
for(j = 0; j <= 16; j++){
printf("%4d", firstfloor[i][j].directions);
}
printf(" |\n");
}
/* The next for loop prints out the lower edge */
for (lower = 0; lower < 18; lower++){
printf("----");
}
printf("-\n");
/*
printf("SECOND FLOOR");
// printf("printfloor is running");
printf("\n-----");
// The next for loop prints out the upper edge
for (upper = 0; upper < 17; upper++){
printf("----");
}
printf("\n");
//The next for loop prints out the floor, every element is 4 digits wide
for (i = 0; i <= 16; i++){
printf("|");
for(j = 0; j <= 16; j++){
printf("%4d", secondfloor[i][j].directions);
}
printf(" |\n");
}
// The next for loop prints out the lower edge
for (lower = 0; lower < 17; lower++){
printf("----");
}
printf("-----\n");
*/
}
prog2.c
#include <stdio.h>
#include <stdlib.h>
#include "motion.h"
//#include "printbothfloors.h"
int main(){
int row = 0;
int column = 0;
initializeArrays();
// printbothfloors();
/*
for (row=0; row < 17; row++){
for (column=0; column < 17; column++){
// firstfloor[row][column].directions=0;
// secondfloor[row][column].directions=5;
//firstfloor[i][j].isRobotHere=0;
//secondfloor[i][j].isRobotHere=0;
//firstfloor[i][j].isMultipleDirections=0;
//secondfloor[i][j].isMultipleDirections=0;
}
}
firstfloor[6][0].directions=5;
firstfloor[7][0].directions=5;
firstfloor[8][0].directions=14;
firstfloor[9][0].directions=5;
firstfloor[10][0].directions=6;
firstfloor[11][1].directions=6;
firstfloor[12][2].directions=6;
firstfloor[13][3].directions=15;
firstfloor[14][4].directions=6;
firstfloor[15][5].directions=6;
firstfloor[16][6].directions=7;
firstfloor[16][7].directions=7;
firstfloor[16][8].directions=7;
firstfloor[16][9].directions=7;
firstfloor[16][10].directions=8;
firstfloor[15][11].directions=8;
firstfloor[14][12].directions=8;
firstfloor[13][13].directions=16;
firstfloor[12][14].directions=8;
firstfloor[11][15].directions=8;
firstfloor[10][16].directions=1;
firstfloor[9][16].directions=1;
firstfloor[8][16].directions=1;
firstfloor[7][16].directions=1;
firstfloor[6][16].directions=2;
firstfloor[5][15].directions=2;
firstfloor[4][14].directions=2;
firstfloor[3][13].directions=17;
firstfloor[2][12].directions=2;
firstfloor[1][11].directions=2;
firstfloor[0][10].directions=3;
firstfloor[0][9].directions=3;
firstfloor[0][8].directions=13;
firstfloor[0][7].directions=3;
firstfloor[0][6].directions=4;
firstfloor[1][5].directions=4;
firstfloor[2][4].directions=4;
firstfloor[3][3].directions=18;
firstfloor[4][2].directions=4;
firstfloor[5][1].directions=4;
firstfloor[1][8].directions=5;
firstfloor[2][8].directions=5;
firstfloor[3][8].directions=5;
firstfloor[4][8].directions=19;
firstfloor[5][8].directions=5;
firstfloor[6][8].directions=5;
firstfloor[7][8].directions=5;
firstfloor[8][8].directions=14;
firstfloor[9][8].directions=5;
firstfloor[10][8].directions=5;
firstfloor[11][8].directions=5;
firstfloor[12][8].directions=19;
firstfloor[13][8].directions=5;
firstfloor[14][8].directions=5;
firstfloor[15][8].directions=5;
firstfloor[8][1].directions=7;
firstfloor[8][2].directions=7;
firstfloor[8][3].directions=7;
firstfloor[8][4].directions=20;
firstfloor[8][5].directions=7;
firstfloor[8][6].directions=7;
firstfloor[8][7].directions=7;
firstfloor[8][9].directions=7;
firstfloor[8][10].directions=7;
firstfloor[8][11].directions=7;
firstfloor[8][12].directions=20;
firstfloor[8][13].directions=7;
firstfloor[8][14].directions=7;
firstfloor[8][15].directions=7;
firstfloor[7][4].directions=11;
firstfloor[9][4].directions=11;
firstfloor[4][7].directions=12;
firstfloor[4][9].directions=12;
firstfloor[12][7].directions=12;
firstfloor[12][9].directions=12;
firstfloor[7][12].directions=11;
firstfloor[9][12].directions=11;
firstfloor[4][4].directions=2;
firstfloor[4][6].directions=7;
firstfloor[4][10].directions=3;
firstfloor[4][12].directions=8;
firstfloor[6][4].directions=5;
firstfloor[6][12].directions=5;
firstfloor[10][4].directions=1;
firstfloor[10][12].directions=1;
firstfloor[12][4].directions=4;
firstfloor[12][6].directions=7;
firstfloor[12][10].directions=3;
firstfloor[12][12].directions=6;
firstfloor[11][0].directions=0;
firstfloor[5][16].directions=0;
*/
// for (i = 0; i < 17; i++){
// firstfloor
// printbothfloors();
// pbf_entrypoints();
/*
row = 0;
column = 0;
while (i < 17){
firstfloor[i][8] = 1;
secondfloor[i][8] = 1;
firstfloor[8][i] = 1;
secondfloor[8][i] = 1;
i++;
}
*/
printf("Function got here");
printbothfloors();
return 0;
}
Valid indices for Type arr[N] are between 0 and N-1.
This goes for any Type and for any number of dimensions.
Indexes in C start at 0, so if you declare an array to have 16 elements, valid indexes start from 0 and end at 15.
16 boxes, numbered 0 to 15:
---------------------------------------------------------------------------------
| 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 |
---------------------------------------------------------------------------------
Accessing an element outside of this range (e.g. by using the index -1 or 16) invokes undefined behaviour.
Both the first answers are correct, I'll add some examples from your code.
firstfloor[6][16].directions
firstfloor[x][y], has been initialized as firstfloor[16][16] meaning the values x and y can can from 0 to 15. The numbering system is 0 based, a hang-over from Java's C language ancestry. 0-15 covers 16 elements.
/*The next for loop prints out the floor, every element is 4 digits wide */
for (i = 0; i <= 16; i++){
printf("|");
for(j = 0; j <= 16; j++){
printf("%4d", firstfloor[i][j].directions);
}
printf(" |\n");
}
The two for loops, start correctly at i = 0 and should only continue while i < 16 or if you prefer i <= 15.
At the moment I'm personally jumping between Java, C and Python, so someone might correct me in this, but calling firstfloor[6][16] if it does not generate an array out-of-bounds error, will give you the same result as firstfloor[7][0]. This might explain why you are observing the first columns being confused with the last.
Good luck :-)