Passing a matrix using pointers in C - c

Compiles without warnings or errors, just the following code crashes when I try to either read from or write to the matrix constValues.
It needs to be passed to another function to be read from also; the function createOutputLine.
How can I point to the data held correctly so it can be modified and read from? Thanks.
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <stdbool.h>
void createOutputFile(FILE*, int, char**);
char createOutputLine(int, int, char*, char**);
int main(int argc, char **argv) {
int removeComments = 0;
FILE *file;
if (argc > 1 && strcmp(argv[1], "-i") == 0) {
if (argc > 2) {
if (!(file = fopen(argv[2], "r"))) {
printf("Error: file not found");
return -1;
}
}
else {
printf("Error: no file specified");
return -1;
}
}
else {
printf("Error: command requires -i");
return -2;
}
createOutputFile(file, argc, argv);
fclose(file);
}
void createOutputFile(FILE *file, int argc, char **argv) {
fseek(file, 0, SEEK_SET);
char *data = (char*)malloc(2000);
FILE *header;
char name[20];
char *token = strtok(argv[2], ".");
strcpy(name, strcat(token, ".o"));
FILE *output = fopen(name, "w");
char constNames[10][15];
char **constValues[10][10];
int constsStored = 0;
while (fgets(data, 2000, file) != NULL) {
for (int i = 0; i < strlen(data); i++) {
int c = i;
bool linePrinted = false;
if (data[i] == '#' && data[i + 1] == 'd') {
for (c = i; c <= i + 7; c++) {
data[c] = '\0';
} int ch = 0;
while (data[c] != ' ') {
constNames[constsStored][ch] = data[c];
data[c] = '\0';
ch++;
c++;
} ch = 0;
while (data[c] != '\n') {
**constValues[constsStored][ch] = data[c]; //this line crashes
data[c] = '\0';
ch++;
c++;
}
if (data[c] == '\n') data[c] = '\0';
constsStored++;
}
for (int ch = 0; ch <= constsStored; ch++) {
if (data[i] == constNames[ch][0]) {
int ch2 = i + 1;
int ch3 = 1;
bool isConst = false;
while (data[ch2] != ' ') {
if (data[ch2] == constNames[ch][ch3] && isConst == false) isConst = true;
ch2++;
ch3++;
}
if (isConst || data[i + 1] == ' ') {
char line[200];
line[200] = createOutputLine(i, ch, data, **constValues);
fprintf(output, "%c", line[200]);
linePrinted = true;
}
}
}
if (!linePrinted)
fprintf(output, "%c", data[i]);
}
}
fclose(output);
free(data);
}
char createOutputLine(int i, int constElem, char *data, char **constValues) {
int ch = i;
int ch2 = 0;
char temp[200];
while (data[ch] != '\n' && data[ch] != ' ' && data[ch] != ';') {
temp[ch2] = data[ch];
printf("%c", data[ch]);
ch++;
ch2++;
}
char line[200];
ch2 = 0;
for (ch = i; ch <= sizeof(data); ch++) {
line[ch2] = data[ch];
ch2++;
}
for (ch = 0; ch <= 10; ch++) {
line[ch2] = constValues[constElem][ch];
ch2++;
}
for (ch = 0; ch <= sizeof(temp); ch++) {
line[ch2] = temp[ch];
ch2++;
}
line[ch2 + 1] = '\n';
return line[200];
}

A pointer shall point to an object before it can be derefenced. Full stop.
char **constValues[10][10]; just declares an 2D array of pointers to pointers to characters. And as it is an automatic array (neither statically nor dynamically allocated), its pointers are just uninitialized.
When you late use **constValues[constsStored][ch] = data[c];, you try to dereference an uninitialized pointer which is explicitely Undefined Behaviour. You are lucky to get an immediate crash, because UB consequences can be apparently unrelated problems.
The normal way is to declare arrays of objects, and use the addresses of those objects for pointers.
That's not all: C arrays are not first class citizens. You cannot assign to array, nor return it from a function. So this is plain wrong:
char line[200];
line[200] = createOutputLine(i, ch, data, **constValues);
It just assigns the unique character returned by the function past the end of the array!
So is this:
char line[200];
...
return line[200];
It does not return an array (C does not allow it) but the value of the byte that happens to live past the array.
I am sorry, but there are too many errors for me to fix them is such a long program.
You may find C hard and ask for help. But build small code containing only what you want to work on. And only when those small pieces work correctly, try to assemble them in a larger program.

Related

How to parse bigger amount of words?

I have a program, which receives filename as an input, saves file contents into 2d char array and then outputs words. It works absolutely fine for about 400 words, but then, when I add more words, it crashes. Debugging showed that i am trying to access unused address, and I don't understand how is that possible considering that previous tests with lesser amount of words were successful.
The question is: what am i missing here?
FILE: functions.c
#include "Lab10.h"
#include <stdio.h>
#include <malloc.h>
char** parser(char* filename) {
FILE* fp;
fp = fopen(filename, "r");
char** str = (char**)calloc(N, sizeof(char*) * N);
if (!str)
{
printf("\n Allocation error");
return NULL;
}
char ch;
int space = 0, words = 0;
for (int i = 0; !feof(fp); i++) // Memory allocation
{
ch = fgetc(fp);
if (!is_ch(ch))
{
if (i != space)
{
if (!(str[words] = (char*)calloc(i - space, sizeof(char) * (i - space))))
{
printf("\n Allocation error");
return NULL;
}
words++;
}
while (!is_ch(ch) && !feof(fp))
{
ch = fgetc(fp);
i++;
}
if(!feof(fp))
fseek(fp, -(int)sizeof(char), 1);
i--;
space = i;
}
}
fseek(fp, 0, SEEK_SET);
for (int i = 0; i < words; i++) // Copying words into 2d array
{
while (!is_ch(fgetc(fp)));
if (!feof(fp))
fseek(fp, -(int)sizeof(char), 1);
int j = 0;
do {
if (((fscanf(fp, "%c", &str[i][j])) != 1))
break;
j++;
} while (is_ch(str[i][j-1]) && !feof(fp));
}
return str;
}
int is_ch(char ch)
{
return ((ch >= 'A' && ch <= 'Z') || (ch >= 'a' && ch <= 'z'));
}
FILE: main.c
#define _CRT_SECURE_NO_WARNINGS
#include "Lab10.h"
#include <stdio.h>
#include <stdlib.h>
int main() {
char* filename = (char*)malloc(sizeof(char) * N);
if (!scanf("%s", filename) || filename == 0)
{
printf("\n Incorrect filename input");
return -1;
}
char** str = parser(filename);
printf("\n Contents of .txt file:");
for (int i = 0; str[i] != NULL; i++) {
printf("\n\t%d) ", i+1);
for (int j = 0; is_ch(str[i][j]); j++) {
printf("%c", str[i][j]);
}
}
return 0;
}
This answer was posted as a reply to one of the comments below the question itself. I tried writing readWord function, which recieves filepointer, reads one word and then returns pointer to the resulting array - that's eases the procedure, making it less complex. It works almost like fgets(), but it reads till non-character, instead of a newline
readWord function itself:
char* readWord(FILE* fp) {
char ch = 0;
while (!is_ch(ch))
{
ch = fgetc(fp);
if (ch == EOF || !ch)
return NULL;
}
int size = 1;
while (is_ch(ch))
{
if ((ch = fgetc(fp)) == EOF || !ch)
break;
size++;
}
fseek(fp, -(size * (int)sizeof(char)), 1);
if (ch != EOF || !ch)
size--;
char* word = (char*)calloc(size, sizeof(char) * size + 1);
if (!word)
{
printf("\n Allocation error.");
return NULL;
}
for (int i = 0; i < size; i++)
word[i] = fgetc(fp);
word[size] = '\0';
return word;
}
That's how i use it in main():
FILE* fp = fopen("test.txt", "r");
char* word;
while ((word = readWord(fp)) != NULL)
{
for (int i = 0; word[i] != '\0'; i++)
printf("%c", word[i]);
printf(" ");
}
Is there is anything i need to improve here? It works fine, but is it possible to somehow make it better?

Why is my solution for Super Reduced String(HackerRank) question working fine on my computer but giving wrong answers in HackerRank?

When I run my code in Hackerrank it fails 6/16 test cases but when I try the same test cases on my computer it works fine.
This is the code that I run on my computer:(I use Clion as ide and the latest MinGW as compiler.)
I initialize the string with one of the test cases that fail on HackerRank.
#include <string.h>
#include <stdio.h>
//#include <stdlib.h>
char* superReducedString(char* s);
int contain(char *S,char find);
void copyWithout(char *S,char *T,char trash);
int countWithout(char *S,char trash);
int findSize(char *S);
void fillString(char *S,char filler);
int main(){
char s[] = {"ppffccmmssnnhhbbmmggxxaaooeeqqeennffzzaaeeyyaaggggeessvvssggbbccnnrrjjxxuuzzbbjjrruuaaccaaoommkkkkxx"};
char *result = superReducedString(s);
printf("%s",result);
}
int findSize(char *S){
int i = 0;
while(*(S+i) != '\0'){
i++;
}
return i;
}
void fillString(char *S,char filler){
int i = 0;
while(*(S+i) != '\0'){
*(S+i) = filler;
i++;
}
}
void copyWithout(char *S,char *T,char trash){
fillString(T,'0');
int i = 0;
int count = 0;
while(*(S+i) != '\0'){
if(*(S+i) != trash){
*(T+count) = *(S+i);
count++;
}
i++;
}
}
int countWithout(char *S,char trash){
int i = 0;
int count = 0;
while(*(S+i) != '\0'){
if(*(S+i) != trash){
count++;
}
i++;
}
return count;
}
int contain(char *S,char find){
int i = 0;
int flag = 0;
while(*(S+i) != '\0'){
if(*(S+i) == find){
flag = 1;
}
i++;
}
return flag;
}
char* superReducedString(char* s){
static char empty[] = "Empty String";
static char result[1024];
int flag = 1;
char temp[findSize(s)];
fillString(temp,'0');
int i,j;//Loop variable.
i = 0;
while(*(s + i) != '\0'){
j = 0;
//Checking if adjacent numbers is same. If it is changing them to '0'.
while(s[j] != '\0') {
if (s[j] == s[j + 1]) {
*(s + j) = '0';
*(s + j + 1) = '0';
}
j++;
}
if(contain(s,'0') == 0){ //If there is no zero in original string that means nothing changed.
return s;
}else{
copyWithout(s,temp,'0');//If there are zeros in it, copy it to a temp char array without zeros.
}
strcpy(s,temp);//Copy temp to s again for swapping.
i++;
}
int count = countWithout(s,'0'); //Calculate the size of original string without zeros.
char finalString[count];//Initialize a new string with the calculated size.
copyWithout(s,finalString,'0'); //Copy original string to finalString without zeros to obtain a clear zeroless string.
strcpy(result,finalString);//copy finalstring to static result string to return it.
i = 0;
while(*(result+i) != '\0'){ //Check if result string consists of zeroes. If it is code will return empty string.
if(*(result+i) != '0'){
flag = 0;
}
i++;
}
if(flag == 0){
return result;
}else{
return empty;
}
}
and this is the code that I run on HackerRank:
#include <assert.h>
#include <ctype.h>
#include <limits.h>
#include <math.h>
#include <stdbool.h>
#include <stddef.h>
#include <stdint.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
char* superReducedString(char* s);
int contain(char *S,char find);
void copyWithout(char *S,char *T,char trash);
int countWithout(char *S,char trash);
int findSize(char *S);
void fillString(char *S,char filler);
char* readline();
int main()
{
FILE* fptr = fopen(getenv("OUTPUT_PATH"), "w");
char* s = readline();
char* result = superReducedString(s);
fprintf(fptr, "%s\n", result);
fclose(fptr);
return 0;
}
char* readline() {
size_t alloc_length = 1024;
size_t data_length = 0;
char* data = malloc(alloc_length);
while (true) {
char* cursor = data + data_length;
char* line = fgets(cursor, alloc_length - data_length, stdin);
if (!line) {
break;
}
data_length += strlen(cursor);
if (data_length < alloc_length - 1 || data[data_length - 1] == '\n') {
break;
}
alloc_length <<= 1;
data = realloc(data, alloc_length);
if (!data) {
data = '\0';
break;
}
}
if (data[data_length - 1] == '\n') {
data[data_length - 1] = '\0';
data = realloc(data, data_length);
if (!data) {
data = '\0';
}
} else {
data = realloc(data, data_length + 1);
if (!data) {
data = '\0';
} else {
data[data_length] = '\0';
}
}
return data;
}
int findSize(char *S){
int i = 0;
while(*(S+i) != '\0'){
i++;
}
return i;
}
void fillString(char *S,char filler){
int i = 0;
while(*(S+i) != '\0'){
*(S+i) = filler;
i++;
}
}
void copyWithout(char *S,char *T,char trash){
fillString(T,'0');
int i = 0;
int count = 0;
while(*(S+i) != '\0'){
if(*(S+i) != trash){
*(T+count) = *(S+i);
count++;
}
i++;
}
}
int countWithout(char *S,char trash){
int i = 0;
int count = 0;
while(*(S+i) != '\0'){
if(*(S+i) != trash){
count++;
}
i++;
}
return count;
}
int contain(char *S,char find){
int i = 0;
int flag = 0;
while(*(S+i) != '\0'){
if(*(S+i) == find){
flag = 1;
}
i++;
}
return flag;
}
char* superReducedString(char* s){
static char empty[] = "Empty String";
static char result[1024];
int flag = 1;
char temp[findSize(s)];
fillString(temp,'0');
int i,j,k;//Loop variable.
i = 0;
while(*(s + i) != '\0'){
j = 0;
while(s[j] != '\0') {
if (s[j] == s[j + 1]) {
*(s + j) = '0';
*(s + j + 1) = '0';
}
j++;
}
if(contain(s,'0') == 0){
return s;
}else{
// printf("temp0 = %s s0 = %s\n",temp,s);
copyWithout(s,temp,'0');
// printf("temp1 = %s s1 = %s\n",temp,s);
}
//printf("%s\n",temp);
strcpy(s,temp);
i++;
}
int count = countWithout(s,'0');
char finalString[count];
copyWithout(s,finalString,'0');
strcpy(result,finalString);
i = 0;
while(*(result+i) != '\0'){
if(*(result+i) != '0'){
flag = 0;
}
i++;
}
if(flag == 0){
return result;
}else{
return empty;
}
}
The only difference is main function and the functions that HackerRank uses for getting input.
I don't know if this helps but sometimes my code can give wrong answers for same input.
What I mean is:
input = "acdqglrfkqyuqfjkxyqvnrtysfrzrmzlygfveulqfpdbhlqdqrrqdqlhbdpfqluevfgylzmrzrfsytrnvqyxkjfquyqkfrlacdqj"
While it should give "acdqgacdqj" as answer, it gives "acdqgacdqjÑ"
The last char randomly changes.
But for other inputs no matter how many times I run it it gives the correct answer on my computer.
char temp[findSize(s)]; fillString(temp,'0'); is invalid. In fillString you iteratate until the element is equal to '\0'. temp is uninitialized - you can't expect it to have any certain value (and even reading an uninitialized value is undefined behavior).
In char finalString[count]; count is too small - it doesn't account for zero terminating character. copyWithout(s,finalString,'0'); is not copying zero terminating character. Which results in strcpy(result,finalString); accessing array out-of-bounds when searching for.... zero terminating character.
When working with C string you usually see a magical + 1 everywhere in the code.
Advices:
Prefer not use variable length arrays (arrays where the size expression is not a constant expression). Prefer using dynamic allocation.
findSize is just strlen...
fillString is just memset(string, value, strlen(string));
When using a compiler, always enable all options. When using gcc, you could use gcc -g -Wall -Wextra -fsanitize=address sourcefile.c - sanitize will allow to really fast find all out-of-bounds accesses on stack variables. Use valgrind to find dynamic allocation leaks.
I advise to change order of argument in copyWithout to (destination, source, fill) so it's the same as strcpy(destination, source) - ie. destination is first.
Seems like fixing parts of code to:
char* superReducedString(char* s){
...
// char temp[findSize(s)];
char temp[findSize(s) + 1];
// fillString(temp,'0');
memset(temp, '0', findSize(s));
temp[findSize(s)] = '\0';
...
char finalString[count + 1];//Initialize a new string with the calculated size.
memset(finalString, '0', count);
finalString[count] = '\0';
}
is enough for me to -fsanitize to stop erroring.
I have no idea where exactly your bug is, but it is quite clear from the output that you are using uninitialised memory. And on your computer, that uninitialised memory contains a zero by pure coincidence, and on the computer used for the test it doesn't.
Generally if you have a problem "it works on computer A but not on computer B" then very often undefined behaviour in your code is the answer, and here it is most like uninitialised memory.

Segmentation fault error while working with command-line argument, C program

I am working on a program that decodes a message provided as a morse code as a command line argument. If the provided string contains 7 spaces it means that I need to print a space to the decoded string. Currently, my program shows all correct letters, but I cannot overcome the issue with locating 7 spaces in a string and printing one space instead.
Here is the issue and how I pass the encoded string:
e3r2p2% ./morse "..- -.. .-. .. .-- --. ..- ....."
udriwg zsh: segmentation fault ./morse "..- -.. .-. .. .-- --. ..- ....."
Program works fine and prints all the decrypted letters (udriwg) till it meets the first (or maybe second) space out of 7;
Here is the code of my program:
int find_morse(const char *s);
bool count_spaces(char *s);
bool is_valid_char(char *s);
bool is_valid_params(int argc, char *argv[]);
void mx_printchar(char c);
void mx_printerr(const char *s);
int mx_strcmp(const char *first, const char *second);
char* mx_strcpy(char* dest, const char* source);
int mx_strlen(const char *s)
int main(int argc, char *argv[]){
if (!is_valid_params(argc, argv)){
mx_printerr("usage: ./morse \"str\"\n");
exit(1);
}
char *output[27]= {"a","b","c","d","e","f","g","h","i","j","k","l","m","n","o",
"p","q","r","s","t","u","v","w","x","y","z","."};
char *copy;
copy = malloc(mx_strlen(argv[1]) + 1);
if (copy != NULL) {
mx_strcpy(copy, argv[1]);
} else {
mx_printerr("usage: ./morse \"str\"\n");
exit(1);
}
int buff = 0;
char temp[5];
int index;
int space = 0;
int j;
for (int i = 0, n = mx_strlen(copy); i < n; i++) {
if (copy[i] == '.' && copy[i + 1] == '.' && copy[i + 2] == '.'
&& copy[i + 3] == '.' && copy[i + 4] == '.') {
mx_printchar(*output[26]);
}
if (copy[i] != ' ') {
temp[buff] = copy[i];
buff++;
}
if (copy[i] == ' ') {
index = find_morse(temp);
//print letter converted from morse to ENG
mx_printchar(*output[index]);
j = i;
while (copy[j] == ' ') {
space++;
j++;
if (space == 7) {
mx_printchar(' ');
space = 0;
break;
}
}
temp[buff] = '\0';
buff = 0;
for(int j = 0; j <4; j++) {
temp[j] ='\0';
}
}
}
mx_printchar('\n');
free(copy);
return 0;
}
int find_morse(const char *s) {
char *input[27]={".-","-...","-.-.","-..",".","..-.","--.", "....", "..",
".---","-.-",".-..","--","-.", "---",".--.","--.-",".-.",
"...","-","..-", "...-",".--","-..-","-.--","--..","....."};
for (int i = 0; i < 27; i++) {
if (mx_strcmp(s, input[i]) == 0) {
return i;
}
}
return -1;
}
bool count_spaces(char *s) {
int space = 0;
for (int i = 0; s[i] != '\0'; i++) {
if (s[i] == ' ') {
space++;
if (s[i + 1] != ' ') {
if (space != 1 && space != 7) {
return false;
} else {
space = 0;
}
}
}
}
return true;
}
bool is_valid_char(char *s){
int i = 0;
while (s[i] != '\0'){
if (s[i] != ' ' && s[i] != '-' && s[i] != '.') {
return false;
}
i++;
}
return true;
}
bool is_valid_params(int argc, char *argv[]){
if (argc != 2) {
return false;
}
if (!is_valid_char(argv[1])){
return false;
}
if (!count_spaces(argv[1])){
return false;
}
return true;
}
void mx_printchar(char c){
write(1, &c, 1);
}
void mx_printerr(const char *s){
write(2, s, mx_strlen(s));
}
int mx_strcmp(const char *first, const char *second) {
while (*first) {
if (*first != *second) break;
first++;
second++;
}
return *(const unsigned char*)first - *(const unsigned char*)second;
}
char* mx_strcpy(char* dest, const char* source){
if (dest == NULL) {
return NULL;
}
char *ptr = dest;
while (*source != '\0') {
*dest = *source;
dest++;
source++;
}
*dest = '\0';
return ptr;
}
int mx_strlen(const char *s){
int length = 0;
while (s[length] != '\0'){
length++;
}
return length;
}
I believe the issue starts here:
j = i;
while (copy[j] == ' ') {
space++;
j++;
if (space == 7) {
mx_printchar(' ');
space = 0;
break;
}
}
Program prints the decripted letter, then tries to check those spaces but somethis go wrong.
I'd appreciate any suggestions as how to resolve it
When using a program from command line you always put a space as a delimiter between your program name and the parameters and other parameters. So, in your case if there are 7 spaces then a extra space will be added as delimiter between your program name and 7 spaces making it to 8 spaces like this ./morse. So, I'd rather suggest to use any other symbol like ! or any other symbol.
Now for the Segmentation fault error I think that's not happening because of that space counter section that's happening because of this line copy = malloc(mx_strlen(argv[1]) + 1); as you must know that malloc() returns a void * to the allocated space.Now here I don't know what those mx_ functions are. Also some questions popped out in my mind while I was debugging your code those answers are included in the modified code(at somewhat satisfying level) i'm putting below. Please answer them in comments.
Code:
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
/* Questions
** 1. What are these mx_ functions?
** 2. What is is_valid_params?
** 3. Can't we define *output[27] as *output = "abcdef...z." ?
** 4. In " for(j = 0; j < 4; j++) temp[j] ='\0'; " why j < 4 instead of j < 5?
*/
// Function Prototype
int find_morse(const char *s);
//main()
int main(int argc, char *argv[])
{
char *output[27]= {"a","b","c","d","e","f","g","h","i","j","k","l","m","n","o","p","q","r","s","t","u","v","w","x","y","z","."};
char *copy, temp[5];
int buff = 0, index, space = 0, i, j, n;
// Checking for valid parameters.
if (argc != 2)
{
printf("usage: ./morse \"Invalid Parameters.\"\n");//mx_printerr("usage: ./morse \"str\"\n");
exit(1);
}
copy = (char *)malloc(strlen(argv[1]) + 1);
if (copy != NULL)
strcpy(copy, argv[1]);
else
{
printf("usage: ./morse \"Memory Unavailable.\"\n");
exit(2);
}
for (i = 0, n = strlen(copy); i < n; i++)
{
if (copy[i] == '.' && copy[i + 1] == '.' && copy[i + 2] == '.' && copy[i + 3] == '.' && copy[i + 4] == '.')
putchar(*output[26]);
if (copy[i] != ' ')
temp[buff++] = copy[i];
else
{
j = i;
while (copy[j] == ' ')
{
space++;
j++;
if (space == 7)
{
putchar(' ');
space = 0;
break;
}
}
temp[buff] = '\0';
buff = 0;
// Emptying temp[5]
//for(j = 0; j < 5; j++)
temp[0] = '\0'; // Rest elements will automatically be assigned to \0 by compiler.
}
if(index = find_morse(temp) != -1)
putchar(*output[index]); //print letter converted from morse to ENG
}
putchar('\n');
free(copy);
return 0;
}
int find_morse(const char *s)
{
char *input[27] = {".-","-...","-.-.","-..",".","..-.","--.", "....", "..",".---","-.-",".-..","--","-.", "---",".--.","--.-",".-.","...","-","..-", "...-",".--","-..-","-.--","--..","....."};
for (int i = 0; i < 27; i++)
{
if (strcmp(s, input[i]) == 0)
return i;
}
return -1;
}
Here in this program I didn't understand that how the user will enter morse code one by one or as a string. Because, when matching those morse code entries, I'm sure a logical error would occur.

Copying contents from one array to another in c

Here is an outline of what my program is suppose to do so far.
While there are more words in inputLine:
take next word from the inputLine
if next word fits in the inputline2
add the nextword to inputline2 (doesn't work and maybe not needed)
add the inputline2 to outputBuffer (doesn't work)
format outputBuffer
Otherwise:
write the outputBuffer to the output file
empty out the outputBuffer (put \0 in position 0)
No matter what I try though, the outputline2 and/or outputBuffer never copy the contents of the inputline properly. The only reason I have inputline2 is because I was originally using fgets and putting the contents from a line in a text file into inputline. However, since my array length is suppose to be 40, it would always cut some of the words in the original line in half. If this could be avoided somehow I wouldn't even need inputline2. Either way, in both cases, the contents from word (which is just a single word from the original inputline) won't ever copy properly.
void format(FILE *ipf, FILE *outf)
{
char inputline[80];
char outputBuffer[MaxOutputLine];
char word[MaxOutputLine];
while(fgets(inputline, 80, ipf) != NULL)
{
int pos = 0;
int i;
int j = 0;
char inputline2[MaxOutputLine] = {'\0'};
while(pos != -1)
{
i=0;
pos = nextword(inputline, word, pos);
if(strlen(word) <= (40 - strlen(inputline2)))
{
while(i < strlen(word))
{
inputline2[j] = word[i];
i++;
j++;
}
j++;
printf("%s", inputline2);
}
}
}
}
int nextword(char *inputline, char *word, int pos1) //takes a word beginning from pos and puts it in word and re\
turns the new position of beginning of the next word
{
int pos2 = 0;
if(inputline[pos1] == '\0')
{
return -1;
}
if(inputline[pos1] == ' ')
{
while(inputline[pos1] == ' ')
{
pos1++;
}
}
else
{
while(inputline[pos1] != ' ' && inputline[pos1] != '\n' && inputline[pos1] != '\0')
{
word[pos2] = inputline[pos1];
pos1++;
pos2++;
}
if(pos1 == '\n' || pos1 == '\0')
{
return -1;
}
pos1++;
}
word[pos2]='\0';
return pos1;
}
As I can't format a code in comment I send it as answer.
First, I guess your nextword() is similar to:
#include <stdio.h>
#include <string.h>
int nextword(char *inputline, char *word, int pos)
{
char *ptr;
int len;
ptr = strchr(&inputline[pos], ' ');
len = ptr ? ptr - &inputline[pos] : strlen(&inputline[pos]);
strncpy(word, &inputline[pos], len);
word[len] = 0;
if (ptr == NULL)
return -1;
return pos + len + 1;
}
To make clear:
#define MaxOutputLine 40
In format() you have missed the line:
if(strlen(word) <= (40 - strlen(inputline2)))
{
while(i < strlen(word))
{
inputline2[j] = word[i];
i++;
j++;
}
inputline2[j] = ' ';/*missed, otherwise undefined, eg. \0 */
j++;
printf("%s", inputline2);
}
Some compilers have problem with long buffers on stack. You can try to add "static" before char buf[] in format().
In nextword() you have the mistake:
if(pos1 == '\n' || pos1 == '\0')
{
return -1;
}
instead of
if(inputline[pos1] == '\n' || inputline[pos1] == '\0')
{
word[pos2]='\0';
return -1;
}
In my opinion nextword() could be simplified to something like that:
int nextword(char *inputline, char *word, int pos1)
{
int pos2 = 0;
while(inputline[pos1] != ' ' && inputline[pos1] != '\n' && inputline[pos1] != '\0')
{
word[pos2] = inputline[pos1];
pos1++;
pos2++;
}
word[pos2]='\0';
if(inputline[pos1] == '\n' || inputline[pos1] == '\0')
{
return -1;
}
return pos1 + 1;
}

Better Way Without Goto?

I have a program where my code uses a goto statement, and I want to get rid of it in a nice way, but I can't seem to find a solution. If goto is the best way, then please let me know. Here is a summary of the code:
//Counts how many times every word appears in a file
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#define NUMWORDS 1000
#define WORDLEN 50
typedef struct
{
char word[WORDLEN + 1];
int num;
} Appearance;
int main(int argc, char *argv[]) {
FILE *readfile;
Appearance *appearlist[NUMWORDS] = {NULL};
char word[WORDLEN + 1];
int i;
//Get a valid filename and open the file, store pointer into readfile
...
char c;
while (c != EOF) {
skip: //Annoying label
//Get a word from readfile, store into word
...
if (word[0] != '\0') {
for (i = 0; i < NUMWORDS && appearlist[i]; i++) {
if (strcmp(appearlist[i] -> word, word) == 0) {
appearlist[i] -> num++;
goto skip; //Annoying goto
}
}
appearlist[i] = (Appearance *) malloc(sizeof(Appearance));
appearlist[i] -> num = 1;
strcpy(appearlist[i] -> word, word);
}
}
//Display results, free memory
...
return 0;
}
The problem is, I want to skip code that is outside of the loop I want to skip from. I would like to not create another variable only designed for this. If you want the full code, click on "Show code snippet."
//Counts how many times every word appears in a file
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#define NUMWORDS 1000
#define WORDLEN 50
#define FILENAMELEN 50
typedef struct
{
char word[WORDLEN + 1];
int num;
} Appearance;
int main(int argc, char *argv[])
{
char filename[FILENAMELEN];
FILE *readfile;
Appearance *appearlist[NUMWORDS] = {NULL};
char word[WORDLEN + 1];
size_t ln;
int i;
if (argc == 2)
strncpy(filename, argv[1], sizeof(filename));
else {
printf("Enter a filename to count appearances from, or just press enter to quit: ");
fgets(filename, FILENAMELEN, stdin);
ln = strlen(filename) - 1;
if (filename[ln] == '\n')
filename[ln] = '\0';
}
while((readfile = fopen(filename, "r")) == NULL) {
if (filename[0] == '\0')
return 0;
printf("Invalid file! Please enter another filename, or just press enter to quit: ");
fgets(filename, FILENAMELEN, stdin);
ln = strlen(filename) - 1;
if (filename[ln] == '\n') filename[ln] = '\0';
}
char c;
while (c != EOF) {
skip:
for (i = 0; (c = getc(readfile)) != EOF && (isalnum(c) || c == '\''); i++) {
if (i >= WORDLEN) {
word[i] = '\0';
printf("\nWarning: word too long (over %d characters), trimming to: %s\n", WORDLEN, word);
while ((c = getc(readfile)) != EOF && (isalnum(c) || c == '\'')) ;
} else {
word[i] = tolower(c);
}
}
word[i] = '\0';
if (word[0] != '\0') {
for (i = 0; i < NUMWORDS && appearlist[i]; i++) {
if (strcmp(appearlist[i] -> word, word) == 0) {
appearlist[i] -> num++;
goto skip;
}
}
appearlist[i] = (Appearance *) malloc(sizeof(Appearance));
appearlist[i] -> num = 1;
strcpy(appearlist[i] -> word, word);
}
}
for (i = 0; i < NUMWORDS && appearlist[i]; i++) {
printf("Word: %s, Appearances: %d\n", appearlist[i] -> word, appearlist[i] -> num);
free(appearlist[i]);
}
return 0;
}
Using goto in this case is often considered acceptable.
Alternatives would be to set a variable so that you can continue in the outer loop after breaking from the inner one, or turning the whole segment that you want to escape from into a separate function, and returning from it instead of using goto.
I'm ignoring any other issues there may be with the code that aren't relevant to the question!
Put everything beginning with the 'if' statement into a separate method (let's call it "process" and replace the goto with return. Then the while-loop becomes:
while (c != EOF) {
//Get a word from readfile, store into word
...
process(...);
}
Sometimes using goto is a hint that code should use a helper function
static bool findword(Appearance *appearlist, size_t size, const char *word) {
for (size_t i = 0; i < size && appearlist[i]; i++) {
if (strcmp(appearlist[i]->word, word) == 0) {
appearlist[i]->num++;
return true;
}
}
return false;
}
while (c != EOF) {
//Get a word from readfile, store into word
...
if (word[0] != '\0') {
if (findword(appearlist, NUMWORDS, word)) {
continue;
}
appearlist[i] = (Appearance *) malloc(sizeof(Appearance));
appearlist[i] -> num = 1;
strcpy(appearlist[i] -> word, word);
}
}

Resources