bitwise macro for getting a bit not working - c

I have a short variable (16 bits) and an index (unsigned char).
I need a macro that returns the indexth in my variable data.
This is what I got:
#define GETBIT(data, index) data & 1 << index
And how I use it:
unsigned char i;
short * twobytes = (short *) calloc(1, sizeof(short));
twobytes = ((char * )buffer + *currentIndex);
while (codeLength != 0)
{
i = GETBIT(code, codeLength--);
*twobytes = SETBIT(*twobytes, *currentBitIndex, i);
(*currentBitIndex)++;
if (*currentBitIndex == 8) {
(*currentIndex)++;
(*currentBitIndex) %= 8;
}
}
For some reason i always equals to 0 in my test cases, where it sometimes should equal 1.
What am I doing wrong and how should I fix it?
Thanks.

Nevermind, thanks, instead of codeLength-- I should've done --codeLength.

Example:
$ cat qq.c
#include <stdio.h>
#define GETBIT(data, index) ((data & (1 << index)) == 0 ? 0 : 1)
int main() {
const int x = 14;
int i;
for (i = 0; i < 32; i++) {
printf("%d ", GETBIT(x, i));
}
printf("\n");
return 0;
}
Run:
$ gcc -Wall qq.c && ./a.out
0 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 0 0 0 0 0 0

Related

I cannot understand core dump cause of fgetc, how can I fix it?

I allocated matrix like this :
prevMatrix = (int**)malloc(sizeof(int) * arraySize[0]);
for (int i = 0; i < arraySize[0]; i++) {
prevMatrix[i] = (int*)malloc(sizeof(int) * arraySize[1]);
}
I checked arraySize[] has normal value. arraySize[0] means row, and arraySize[1] means column.
and I handed over the matrix like this :
void getInputMatrix(FILE* input, int column, int** matrix)
getInputMatrix(input, arraySize[1], prevMatrix);
and then function's body is this :
void getInputMatrix(FILE* input, int column, int** matrix) {
int i = 0, j = 0, c = 0;
while (!feof(input)) {
if (j == column) {
j = 0;
i++;
}
c = fgetc(input);
if (c != EOF && c != 32 && c != 10 && c != 13) {
matrix[i][j] = c - 48;
j++;
}
}
}
this is the example of matrix file:
1 0 0 0 0 0 0
0 1 0 1 0 0 0
0 1 0 0 1 1 0
0 1 1 1 0 0 0
0 0 0 0 1 0 0
0 1 0 0 0 1 0
1 0 1 1 0 0 0
0 1 0 0 0 0 1
It works very well in VS 2019, but it cause segmentation fault in Linux system(Ubuntu).
This program works well in Linux till the column is 6 or 7, but it occurs corrupted size vs. prev_size error or segmentation faultwhen the column overs that number.
Is it a problem of allocation? or fgetc()?
How can I fix it?
For starters, your first level allocation should be using the size of an int pointer rather than an int:
int **prevMatrix = malloc(sizeof(int*) * arraySize[0]);
If those types are different sizes, your original code could have a problem.
And, just as an aside, you should not cast the malloc return value in C, it can cause subtle problems.

Implementing 8-Connectivity Hoshen-Kopelman Algorithm in C

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];
}

Bit mask with 0's in C

I need to build a method in C that will return an int, take 3 ints as parameters. The first and second int are the starting and ending bit position. The third int is a 0 or 1 to determine the type of mask.
For example,
getMask(2, 6, 1);
//Set bits 2 to 6 to 1, set all others to zero
should set the bits 2 through 6 to a 1 and all other bits to zero.
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 0 0
So getMask(2, 6, 1) should return the integer 124.
And getMask(11, 31, 0) (set bits 11 to 31 to 0) should return 2047.
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 1 1 1 1 1
This is what I have currently:
#include <stdio.h>
int getM(int start, int end, int choice) {
if (choice == 1){
return ~(~0 << (end - start + 1)) << (start);
}
else if (choice == 0){
return ~(~1 << (end - start + 1)) << (start);
}
else{
return 0;
}
}
It works when I the choice is 1, but for 0 I am completely lost.
I currently get -2048 for getMask(11, 31, 0).
I know I can use ands and ors, but I cannot figure out how to use them the way I am doing this.
#AnttiHaapala is correct: choice==0 is just the bitwise negation of choice==1 for the same start and end. Therefore (as an MCVE):
#include <stdio.h>
int getM(int start, int end, int choice) {
if (choice == 1){
return ~(~0 << (end - start + 1)) << (start);
}
else if (choice == 0){
return ~getM(start, end, 1); /* Just use what you have, but ~ it */
}
else{
return 0;
}
}
int main() {
printf("2 6 1 %d\n", getM(2,6,1));
printf("11 31 0 %d\n", getM(11,31,0));
}

Adjacency matrix of graph in c ( getting error )

#include <stdio.h>
struct graph{
int v,e,**adj;
};
struct graph* graph_A(){
int u,V,i;
struct graph *g = (struct graph*)malloc(sizeof(struct graph));
if(!g){
printf("error ! ");
exit(1);
}
printf("Enter no of nodes and edges ? ");
scanf("%d, %d",&g->v,&g->e);
g->adj = (int**)malloc(sizeof(int) * sizeof(g->v * g->v ));
for(u = 0 ; u <= g->v ; u++)
for(V = 0 ; V <= g->v ; V++ )
g->adj[u][V] = 0;
printf("Enter connections ? ");
for(i = 0 ; i <= g->e ; i++){
scanf("%d, %d",&u,&V);
g->adj[u][V] = 1;
g->adj[V][u] = 1;
}
return (g);
}
int main(){
graph_A();
}
I am trying to figure out whats the problem but unable to resolve by myself . please help !
i've researched for about 2 hours over this topic unfortunately did't found any
proper help desk.
i am learning data structure and implementing in c and c++ language.
if you could able to advice me
that'd be earnestly kind of you .
thanks !
Slightly fixed your code using vectors. Not the cleaniest way due to non-use of auto loops, bad choices of variables, no boundary check on user input, but at least it works
I have removed the allocations fully, replaced by vectors, much simpler, and the fact is that the code did not change that much.
Note that you don't need a graph class. Everything could have been done in the main itself. If you want to create a class, then provide parameters in the constructor instead of interactive input (do that in the main so others can reuse your class for other purposes), and provide methods to compute, get access to data, size, etc... That will come afterwards.
Which were the main concerns:
malloc of the instance within the constructor itself. Using new would have created an infinite recursion. My advice: read some C++ books
all loops include upper boundary
the pointer of pointer row was allocated but not the rows themselves.
When I rewrote the code, I had 2 compilation errors, one SEGV because of the last upper bound boundary error on the display part, after that it worked right away. C++ rules (when you know your way around)
code:
#include <iostream>
#include <cstdlib>
#include <vector>
using namespace std;
class graph{
int v,e;
vector<vector<int> > adj;
public:
graph();
};
graph::graph()
{
int u,V,i;
cout << "Enter no of nodes and edges ? ";
cin >> v ;
cin >> e ;
adj.resize(v);
for(u = 0 ; u < v ; u++)
{
adj[u].resize(v);
}
for(u = 0 ; u < v ; u++)
for(V = 0 ; V < v ; V++ )
adj[u][V] = 0;
cout << " Enter connections ? ";
for(i = 0 ; i < e ; i++){
cin >> u ;
cin >> V ;
adj[u][V] = 1;
adj[V][u] = 1;
}
for(u = 0 ; u < v ; u++)
{
for(V = 0 ; V < v ; V++ )
{
cout << adj[u][V] << " ";
}
cout << endl;
}
}
int main(){
graph a;
return 0;
}
test:
Enter no of nodes and edges ? 10 3
Enter connections ? 3 4
2 3
4 5
0 0 0 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 0
0 0 1 0 1 0 0 0 0 0
0 0 0 1 0 1 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 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

Bitboard chess programming in C

I have a problem with this piece of code in C.
#include <stdio.h>
#include <stdint.h>
typedef uint64_t bboard;
// Accessing a square of the bitboard
int
get (bboard b, int square)
{
return (b & (1ULL << square));
}
void
print_board (bboard b)
{
int i, j, square;
for (i = 7; i >= 0; i--) // rank => top to bottom
{
for (j = 0; j < 8; j++) // file => left to right
printf ("%d ", get (b, j+8*i) ? 1 : 0);
printf ("\n");
}
}
int
main ()
{
bboard b = 0xffffffffffffffff;
print_board (b);
}
// result that I have
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 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
Ok, why the bitboard is not set with all bit at 1?
For any question please add a comment. Ty :D
get returns an int, but (b & (1ULL << square)) is a uint64_t. When (b & (1ULL << square)) is greater than INT_MAX, the result is undefined; in this case it truncates and returns 0.
If get returns a bboard instead, this works as expected (verified here: http://codepad.org/zEZiJKeR).

Resources