C: search for non-number-character in array - c

I am building a C program which gets a .txt file as an argument. Inside the .txt there are rows of integer numbers, with a space between them, like this:
1 2 3 4
5 6 7 8
...
I am supposed to find out, if a non-integer-character shows up inside the .txt file, like this:
1 2 a 4
...
Since there is no instanceof operator in C, I use an array which contains the numbers from 0-9 and check each character of the .txt-file, if it is either a space or an integer. If it is neither, the program is supposed to exit.
If there are no problems in the file, the program calculates a median and prints the line to stdout.
This is my code:
#include <stdio.h>
#include <ctype.h>
int arrayContains(char value);
int main(int argc, char **argv) {
const int LINESIZE = 255;
if (argc != 2) {
printf("wrong args!");
return -1;
}
char *command1 = argv[1];
FILE *handle = fopen(command1, "r");
if (!handle) {
printf("file not found!");
return -1;
}
int count = 0;
int sum = 0;
int median;
char string[LINESIZE];
while (fgets(string, LINESIZE, handle) != NULL) {
for (int i = 0; i <= sizeof(string) / sizeof(string[0]) - 1; i++) {
printf("%c", string[i]);
if (string[i] == ' ') {
i++;
}
else if (arrayContains(string[i]) == 0) {
count++;
sum += (int)string[i];
}
else {
printf("non-integer-character found!\n");
return -1;
}
}
median = sum / count;
printf("%s\n", string);
printf("%d\n", median);
}
}
int arrayContains(char value) {
const char numbers[10] = {'0', '1', '2', '3', '4', '5', '6', '7', '8', '9'};
for (int i = 0; i <= 9; i++) {
if (numbers[i] == value) {
return 0;
}
}
return -1;
}
Now the output of the program is always the first number of the txt-file and immediately after that: "non-integer-character found!".
Which tells me that the comparison between the element of the string from gets() and an element from the constant "numbers" array inside the arrayContains() function, will always result in a return of -1 from the arrayContains() function.
What am I doing wrong?

You should really check out isdigit C function!
#include <ctype.h>
int hasDigit(const char *s)
{
while (*s) {
if (isdigit(*s++) == 0) return 0;
}
return 1;
}
Code above returns 1 on empty string. Watch out! You can use my function instead of your arrayContains. If you want to use your routine, please check out my "version":
int arrayContains(char value) {
return isdigit(value)?0:-1;
}

There are several problems with your code.
This bit matches a space, but will result in you skipping the following character because your loop will also do i++.
if (string[i] == ' ') {
i++;
}
This is not the right way to turn a digit into a number. What you're getting here is the ASCII value of the character rather than the value of the digit. So for example if you have a '1' you're adding 49 to sum rather than 1.
sum += (int)string[i];
As discussed elsewhere, you're better off using isdigit() to identify if you've got a digit character. You can also use isspace() to test to see if you have a space or '\n' character (it covers all whitespace). Which would make your loop statement a lot less complicated as you can process the whole string and easily handle lines that are longer than the size of your buffer.
This code corrects the problems you have
while (fgets(string, LINESIZE, handle) != NULL) {
for (char *pos=string; *pos!='\0'; pos++) {
printf("%c", *pos);
if (isdigit(*pos)) {
count++;
sum += *pos-'0';
} elseif(!isspace(*pos)) {
printf("non-integer-character found!\n");
return -1;
}
}
median = sum / count;
printf("%s\n", string);
printf("%d\n", median);
}

I seem to have solved it:
#include <stdio.h>
#include <ctype.h>
int arrayContains(char value);
int main(int argc, char **argv) {
const int LINESIZE = 255;
if (argc != 2) {
printf("wrong args!");
return -1;
}
char *command1 = argv[1];
FILE *handle = fopen(command1, "r");
if (!handle) {
printf("file not found!");
return -1;
}
int count = 0;
int sum = 0;
int median;
char string[LINESIZE];
while (fgets(string, LINESIZE, handle) != NULL) {
for (int i = 0; i <= sizeof(string) / sizeof(string[0]) - 1; i++) {
printf("%c\n", string[i]);
if (isspace(string[i]) == 0) {
i++;
}
else if (isdigit(string[i]) == 0) {
count++;
sum += (int)string[i];
}
else {
printf("non-integer-character found!\n");
return -1;
}
}
fgets(string, LINESIZE, handle);
}
median = sum / count;
printf("%s\n", string);
printf("%d\n", median);
}
it now kinda does the job as expected.

You use the operator sizeof() which won't return the length of the String but the memory size of the pointer (a size_t so 8 bytes).
I suggest you to use this for your for loop:
for (int i = 0; string[i] != '\0' && string[i] != '\n'; i++) {
...
}
String in C are just a part of memory, it's just like an array and the only way to get the length is to find the end ('\0')
I also suggest you to directly search into the ASCII table. Characters are just number between 0 and 127. Digits are between 48 and 57, a simple condition does the stuff !
if (string[i] <= 48 || string[i] >= 57) {
...
}

Related

remove the "\n" from string

I'm learning C recently and I didn't learn about pointers so still not allowed to use it.
Note: i'm not allowed to use it or any function from string.h library.
I wrote a function that removes the "\n" from a string.
when I run my program appears to me:
main.c:20:14: warning: comparison between pointer and integer
main.c:80:39: warning: format not a string literal and no format arguments [-Wformat-security]
This is my function:
#include <stdio.h>
#define STRING_SIZE 100
void replace(char str[]){
int i=0;
while(str[i]!='\0'){
if(str[i]=="\n"){
str[i]='\0';
}
}
}
int my_strlen(char s[]) {
int i = 0;
while (s[i] != '\0') {
i++;
}
return i;
}
int remover(char s1[], char s2[], char s3[]) //removes the sustring s2 from string s1 and saved it to s3
{
int i = 0, j, t = 0, found;
while (s1[i])
{
found = 1;//Initilize found to true
for (j = 0; s2[j] != 0; j++) {
if (s1[i + j] != s2[j])
found = 0;//Set not found
}
if (found == 0) {
s3[t] = s1[i];// if not found add char to s3.
t++;
}
else {
i = i + my_strlen(s2) - 1;//if found skip
}
i++;
}
s3[t] = 0;
if (my_strlen(s1) > my_strlen(s3)) {
return 1;
}
return 0;
}
int main() {
char result_string[STRING_SIZE+1], MainString[STRING_SIZE+1], PatternString[STRING_SIZE+1];
printf("Please enter the main string..\n");
fgets(MainString, STRING_SIZE + 1, stdin);
replace(MainString);
printf("Please enter the pattern string to find..\n");
fgets(PatternString, STRING_SIZE + 1, stdin);
replace(PatternString);
int is_stripped = remover(MainString, PatternString, result_string);
printf("> ");
printf(is_stripped ? result_string : "Cannot find the pattern in the string!");
return 0;
}
what's the problem?
You have a lot of problems:
Your function returns a char, which is a single character.
A C-style string has to have a terminating zero byte. You don't put one on helper. So even if you could return it properly, the code that got it would have no way to know how long the string was.
You allocate helper on the stack in replace, so helper stops existing when you return. So where will the returned string be stored?
Just remove the '\n' from the string, in place, modifying the original string.
For instance:
void remove_newline(char str[])
{
int i;
for(i=0; str[i] != 0; ++i)
{
if (str[i] == '\n')
{
str[i] = 0;
break;
}
}
}

returning string in C function [closed]

Closed. This question needs debugging details. It is not currently accepting answers.
Edit the question to include desired behavior, a specific problem or error, and the shortest code necessary to reproduce the problem. This will help others answer the question.
Closed 6 years ago.
Improve this question
I was trying to solve CountAndSay problem at one of the online coding site but I am not able to get why my program is printing NULL. I am sure I am doing some conceptual mistake but not getting it.
Here is my code :
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
char* countAndSay(int A) {
int i,j,k,f,count;
char a;
char *c = (char*)malloc(sizeof(char)*100);
char *temp = (char*)malloc(sizeof(char)*100);
c[0] = 1;c[1] = '\0';
for(k=2; k<=A; k++)
{
for(i=0, j=0; i<strlen(c); i++)
{
a = c[i];
count = 1;
i++;
while(c[i] != '\0')
{
if(c[i]==a)
{
count++;
i++;
}
else if(c[i] != a)
{
i--;
break;
}
else
{
break;
}
}
temp[j] = count;
temp[j+1] = a;
j += 2;
}
*(temp+j) = '\0';
if(k<A)
{
for(j=0; j<strlen(temp); j++)
{
c[j] = temp[j];
}
c[j] = '\0';
}
}
return temp;
}
int main(void) {
// your code goes here
char *c = countAndSay(8);
printf("%s\n",c);
return 0;
}
The idea is not that bad, the main errors are the mix-up of numerical digits and characters as shown in the comments.
Also: if you use dynamic memory, than use dynamic memory. If you only want to use a fixed small amount you should use the stack instead, e.g.: c[100], but that came up in the comments, too. You also need only one piece of memory. Here is a working example based on your code:
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
// ALL CHECKS OMMITTED!
char *countAndSay(int A)
{
int k, count, j;
// "i" gets compared against the output of
// strlen() which is of type size_t
size_t i;
char a;
// Seed needs two bytes of memory
char *c = malloc(2);
// Another pointer, pointing to the same memory later.
// Set to NULL to avoid an extra malloc()
char *temp = NULL;
// a temporary pointer needed for realloc()-ing
char *cp;
// fill c with seed
c[0] = '1';
c[1] = '\0';
if (A == 1) {
return c;
}
// assuming 1-based input, that is: the first
// entry of the sequence is numbered 1 (one)
for (k = 2; k <= A; k++) {
// Memory needed is twice the size of
// the former entry at most.
// (Averages to Conway's constant but that
// number is not usable here, it is only a limit)
cp = realloc(temp, strlen(c) * 2 + 1);
temp = cp;
for (i = 0, j = 0; i < strlen(c); i++) {
//printf("A i = %zu, j = %zu\n",i,j);
a = c[i];
count = 1;
i++;
while (c[i] != '\0') {
if (c[i] == a) {
count++;
i++;
} else {
i--;
break;
}
}
temp[j++] = count + '0';
temp[j++] = a;
//printf("B i = %zu, j = %zu\n",i,j-1)
//printf("B i = %zu, j = %zu\n",i,j);
}
temp[j] = '\0';
if (k < A) {
// Just point "c" to the new sequence in "temp".
// Why does this work and temp doesn't overwrite c later?
// Or does it *not* always work and fails at one point?
// A mystery! Try to find it out! Some hints in the code.
c = temp;
temp = NULL;
}
// intermediate results:
//printf("%s\n\n",c);
}
return temp;
}
int main(int argc, char **argv)
{
// your code goes here
char *c = countAndSay(atoi(argv[1]));
printf("%s\n", c);
free(c);
return 0;
}
To get a way to check for sequences not in the list over at OEIS, I rummaged around in my attic and found this little "gem":
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <errno.h>
#include <limits.h>
char *conway(char *s)
{
char *seq;
char c;
size_t len, count, i = 0;
len = strlen(s);
/*
* Worst case is twice as large as the input, e.g.:
* 1 -> 11
* 21 -> 1211
*/
seq = malloc(len * 2 + 1);
if (seq == NULL) {
return NULL;
}
while (len) {
// counter for occurrences of ...
count = 0;
// ... this character
c = s[0];
// as long as the string "s"
while (*s != '\0' && *s == c) {
// move pointer to next character
s++;
// increment counter
count++;
// decrement the length of the string
len--;
}
// to keep it simple, fail if c > 9
// but that cannot happen with a seed of 1
// which is used here.
// For other seeds it might be necessary to
// use a map with the higher digits as characters.
// If it is not possible to fit it into a
// character, the approach with a C-string is
// obviously not reasonable anymore.
if (count > 9) {
free(seq);
return NULL;
}
// append counter as a character
seq[i++] = (char) (count + '0');
// append character "c" from above
seq[i++] = c;
}
// return a proper C-string
seq[i] = '\0';
return seq;
}
int main(int argc, char **argv)
{
long i, n;
char *seq0, *seq1;
if (argc != 2) {
fprintf(stderr, "Usage: %s n>0\n", argv[0]);
exit(EXIT_FAILURE);
}
// reset errno, just in case
errno = 0;
// get amount from commandline
n = strtol(argv[1], NULL, 0);
if ((errno == ERANGE && (n == LONG_MAX || n == LONG_MIN))
|| (errno != 0 && n == 0)) {
fprintf(stderr, "strtol failed: %s\n", strerror(errno));
exit(EXIT_FAILURE);
}
if (n <= 0) {
fprintf(stderr, "Usage: %s n>0\n", argv[0]);
exit(EXIT_FAILURE);
}
// allocate space for seed value "1" plus '\0'
// If the seed is changed the limit in the conway() function
// above might need a change.
seq0 = malloc(2);
if (seq0 == NULL) {
fprintf(stderr, "malloc() failed to allocate a measly 2 bytes!?\n");
exit(EXIT_FAILURE);
}
// put the initial value into the freshly allocated memory
strcpy(seq0, "1");
// print it, nicely formatted
/*
* putc('1', stdout);
* if (n == 1) {
* putc('\n', stdout);
* free(seq0);
* exit(EXIT_SUCCESS);
* } else {
* printf(", ");
* }
*/
if (n == 1) {
puts("1");
free(seq0);
exit(EXIT_SUCCESS);
}
// adjust count
n--;
for (i = 0; i < n; i++) {
// compute conway sequence as a recursion
seq1 = conway(seq0);
if (seq1 == NULL) {
fprintf(stderr, "conway() failed, probably because malloc() failed\n");
exit(EXIT_FAILURE);
}
// make room
free(seq0);
seq0 = NULL;
// print sequence, comma separated
// printf("%s%s", seq1, (i < n - 1) ? "," : "\n");
// or print sequence and length of sequence, line separated
// printf("%zu: %s%s", strlen(seq1), seq1, (i < n-1) ? "\n\n" : "\n");
// print the endresult only
if (i == n - 1) {
printf("%s\n", seq1);
}
// reuse seq0
seq0 = seq1;
// not necessary but deemed good style by some
// although frowned upon by others
seq1 = NULL;
}
// free the last memory
free(seq0);
exit(EXIT_SUCCESS);
}

C substring two equal characters not matching

am writing a program to find a substring in a string. For some strange reason, during comparison of one character to another, two similar characters do not match. I'm not sure if this is some sort of bug.
Here's the code:
#include <stdio.h>
int first_char_compare(char *str,char a);
int main()
{
int i,j, char_exists,substring;
char string1[] = "jrsfahamfsf";
char string2[] = "ham";
char_exists = first_char_compare(string1,string2[0]);
if(!(char_exists<0))
{
j=char_exists;
for(i=0;string2[i]!='\0';i++)
{
for(;string1[j]!='\0';j++)
{
printf("%c\t%c\n",string2[i],string1[j]);
if(string1[i]==string2[j])
{
substring = 1;
printf("matches\n");
break;
}
else
{
printf("doesn't match\n");
substring = 0;
}
}
if(substring==0)
break;
j++;
}
}
if(substring)
printf("Is a substring\n");
else
printf("Not a substring\n");
return 0;
}
int first_char_compare(char *str,char a)
{
/* Checks if the first character of the substring is in the main string.
* If the character exists, it's index is returned. If it doesn't exist
* -1 is returned.
*/
int i;
for(i=0;str[i]!='\0';i++)
{
if(a==str[i])
return i;
}
return -1;
}
In your loops, string2 is indexed by i, and string1 by j:
for (i = 0; string2[i] != '\0'; i++)
{
for( ; string1[j] != '\0'; j++)
{
But in your comparison, the indices are backwards:
if (string1[i] == string2[j])
correct that to:
if (string1[j] == string2[i])

How to exit scanf loop when there is a space

For example, the user shall put the input like that, "ABC123," but not "ABC 123" or "A BC123."
Here is my code:
unsigned int convert_to_num(char * string) {
unsigned result = 0;
char ch;
//printf("check this one %s\n", string);
while(ch =*string++) result = result * 26 + ch - 'A' + 1;
return result;
}
int main()
{
char input_string[100];
char arr_col[100] = {'\0'};
char arr_row[100] = {'\0'};
int raiseflag;
int started_w_alpha =0;
int digitflag = 0;
while(scanf("%s", &input_string) != EOF) {
int i = 0, j = 0, digarr = 0;
while (i <=5) {
if (input_string[i] == '\0') {printf("space found!");}
if ((input_string[i] >= 'A' && input_string[i] <= 'Z') && (digitflag == 0)) {
started_w_alpha = 1;
arr_col[j] = input_string[i]; j++;
}
//printf("something wrong here %s and %d and j %d\n", arr_holder, i, j);
if (started_w_alpha == 1) {
if (input_string[i] >=48 && input_string[i]<=57){ digitflag = 1; arr_row[digarr] =input_string[i]; digarr++; }
}
i++; if (i == 5) { raiseflag =1; }
}
printf(" => [%d,%s]\n", convert_to_num(arr_col), arr_row);
if (raiseflag == 1) { raiseflag = 0; memset(arr_col, 0, 5); memset(input_string, 0, 5); memset(arr_row, 0, 5); digitflag = 0; started_w_alpha = 0; }
}
return 0;
}
Apparently, \0 doesn't work in my case because I have an array of 5 and user can put 2 chars. I want to exit the loop whenever a space is found in between the characters.
This is the whole code. I added {'\0'} my array because of the extra characters I get when there is less than 5 characters.
Thanks!
Since the index is starting from 0 and input_string[5]; array size is 5, the only valid indexes are from 0 to 4.
but your loop while (i <=5) { go till 5, it is mean you exceed the array.
If you insert 5 characters to the string, the terminating null is the 6th.
Since you exceed the array it written over some other variable. but you still can find it when you check input_string[5]
So if you want to insert 5 characters you array size should be at least 6
char input_string[6];
if you want to check only the first 5 elements you'll have to change the loop to:
while (i < 5) {
and as I wrote in the comment if you find the terminating null, no use to continue the loop, since it contain garbage or leftover from the previous iteration.
Therefor you should break if it found, like this:
if (input_string[i] == '\0') {printf("space found!"); break;}
EDIT
check this program: it use fgets to read the whole input, then search for white spaces.
Note it doesn't trim the input, means it won't remove spaces when thay appear at the beginning or at the end of the input.
#include <ctype.h>
#include <string.h>
#include <stdio.h>
int main()
{
int i ,size;
char input_string[100];
fgets(input_string,100,stdin);
i=0;
size = strlen(input_string);
while (i<size-1){ //enter is also count
if (isspace(input_string[i]))
{
printf("space found!");
break;
}
i++;
}
return 0;
}
EDIT2
Now with a trim, so it will remove leading and ending spaces:
#include <ctype.h>
#include <string.h>
#include <stdio.h>
char* trim(char *input_string)
{
int i=0;
char *retVal = input_string;
i = strlen(input_string)-1;
while( i>=0 && isspace(input_string[i]) ){
input_string[i] = 0;
i--;
}
i=0;
while(*retVal && isspace(retVal[0]) ){
retVal ++;
}
return retVal;
}
int main()
{
int i ,size;
char input_string[100],*ptr;
fgets(input_string,100,stdin);
ptr = trim(input_string);
i=0;
size = strlen(ptr);
while (i<size){
if (isspace(ptr[i]))
{
printf("space found!");
break;
}
i++;
}
return 0;
}

Convert a set of chars to int from a file

I'm reading:
22:5412:99:00 (...)
From a text file using (ch=fgetc(fp)) != EOF because I don't have only those numbers to read.
Identifying a number is easy with if(ch >= 48 && ch <= 57) but the thing is I want to put those numbers 22, 5412 into an array of integers. However when I read a char it reads part of number since each number is char.
It gets 2 (and not 22 like I want to) and in the next iteration reads the other 2. How can I save each set of numbers into it's own integer?
I hope I was clear enough, thanks!
My idea is to read in each char, and if it is a digit append it to a buffer. Whenever we get a non-digit, we just read the contents of the buffer as a string using sscanf, and clear the buffer for the next value.
#include <stdio.h>
#include <stdlib.h>
int read_buffer(char* buffer, int* sz)
{
int ret;
if (*sz==0) return 0;
buffer[*sz]='\0'; //end the string
sscanf(buffer,"%d", &ret); //read the contents into an int
*sz=0; // clear the buffer
return ret;
}
int main()
{
char buffer[1000];
int sz=0;
char ch;
FILE* input=fopen("input.txt","r");
// "input.txt" contains 22:5412:99:00
while ((ch=fgetc(input))!=EOF)
{
int number;
if (isdigit(ch))
{
buffer[sz++]=ch; // append to buffer
}
else
{
printf("Got %d\n",read_buffer(buffer,&sz)); // read contents of buffer and clear it
}
}
if (sz) // check if EOF occured while we were reading a number
printf("Got %d\n",read_buffer(buffer,&sz));
fclose(input);
return 0;
}
You would need to store the numbers as a string or a char* and use atoi to actually convert it to a number. http://www.cplusplus.com/reference/clibrary/cstdlib/atoi/
Assuming your pattern is of the type NN:NNNN:NN:NN, parse on the delimiter, feeding characters into a buffer:
int idx = 0, nIdx = 1;
int firstN, secondN, thirdN, fourthN;
char buf[5];
...
while ((ch=fgetc(fp)) != EOF) {
if (ch != ':') {
buf[idx++] = ch;
}
else {
buf[idx] = '\0';
idx = 0;
switch (nIdx++): {
case 1: firstN = atoi(buf); break;
case 2: secondN = atoi(buf); break;
case 3: thirdN = atoi(buf); break;
}
}
}
buf[idx] = '\0';
fourthN = atoi(buf);
...
I did a full program out of the previous post -- and some testing :-)
#include <ctype.h>
#include <stdio.h>
#include <stdlib.h>
/* fill `array` with at most `siz` values read from the stream `fp` */
/* return the number of elements read */
size_t fillarray(int *array, size_t siz, FILE *fp) {
int ch;
size_t curr = 0;
int advance_index = 0;
while ((ch = fgetc(fp)) != EOF) {
if (isdigit((unsigned char)ch)) {
array[curr] *= 10;
array[curr] += ch - '0';
advance_index = 1;
} else {
if (advance_index) {
advance_index = 0;
curr++;
if (curr == siz) { /* array is full */
break;
}
}
}
}
return curr + advance_index;
}
int main(void) {
int array[1000] = {0};
int n, k;
n = fillarray(array, 1000, stdin);
if (n > 0) {
printf("%d values read:\n", n);
for (k=0; k<n; k++) {
printf(" %d", array[k]);
}
puts("");
} else {
fprintf(stderr, "no data read\n");
}
return 0;
}
And a test run
$ ./a.out
24555:76423 foobar 76235 jgfs(8) jhg x86-64 passw0rd RS232
[CTRL+D]
8 values read:
24555 76423 76235 8 86 64 0 232

Resources