Printing top 10 recurring words in a file - c

Question

Edited question:
Hi guys, my goal is to print the top 10 occurring words in a file, I have managed to get everything to work from reading the file to counting word occurrences and printing it, but when I implement my qsort I get a segfault. I looked over my pointers and they look okay to me, I would appreciate any feedback.
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <ctype.h>
#define MAX 51
struct words
{
char *ch;
int index;
struct words *pNext;
};
struct words* createWordCounter(char *ch)
{
struct words *pCounter = NULL;
pCounter = (struct words*)malloc(sizeof(char));
pCounter->ch = (char*)malloc(strlen(ch)+1);
strcpy(pCounter->ch, ch);
pCounter->index = 1;
pCounter->pNext = NULL;
return pCounter;
}
struct words *pStart = NULL;
char* removePunc(struct words* ch)
{
char *src = ch, *dst = ch;
while (*src)
{
if (ispunct((unsigned char)*src))
{
src++;
}
else if (isupper((unsigned char)*src))
{
*dst++ = tolower((unsigned char)*src);
src++;
}
else if (src == dst)
{
src++;
dst++;
}
else
{
*dst++ = *src++;
}
}
*dst = 0;
}
void addWord(char *word)
{
struct words *pCounter = NULL;
struct words *pLast = NULL;
if(pStart == NULL)
{
pStart = createWordCounter(word);
return;
}
pCounter = pStart;
while(pCounter != NULL)
{
if(strcmp(word, pCounter->ch) == 0)
{
++pCounter->index;
return;
}
pLast = pCounter;
pCounter = pCounter->pNext;
}
pLast->pNext = createWordCounter(word);
}
void printWord(struct words *pCounter)
{
printf("\n%-30s %5d\n", pCounter->ch, pCounter->index);
}
//sort
int compare (const void * a, const void * b){
struct words *A1 = (struct words *)a;
struct words *B1 = (struct words *)b;
return B1->index - A1->index;
/*
if ((A1->count - B1->count) > 0)
return -1;
else if ((A1->count - B2->count) < 0)
return 1;
else
return 0;
*/
}
int main(int argc, char * argv[])
{
struct words *pCounter = NULL;
char temp[MAX];
FILE *fpt;
if(argc == 2)
{
printf("File name is: %s\n",argv[1]);
fpt = fopen(argv[1], "r");
//fail test
if(fpt == NULL)
{
printf("cannot open file, exiting program...\n");
exit(0);
}
//get the data out of the file and insert in struct
int wordCounter = 0;
int i = 0;
int lines = 0;
while((fscanf(fpt, "%s ", &temp)) == 1)
{
removePunc(temp);
addWord(temp);
if(temp == ' ')
i++;
if(temp == '\n')
lines++;
wordCounter++;
}
/*
pCounter = pStart;
while(pCounter != NULL)
{
printWord(pCounter);
pCounter = pCounter->pNext;
}
*/
//sort
qsort(pCounter, wordCounter, sizeof(struct words), compare);
for(int j = 0; i < 10; i++)
{
printWord(pCounter);
}
}
fclose(fpt);
return 0;
}

Answer 1

First temp is already a pointer, so do not include '&' before it in fscanf. Second, don't skimp on buffer size (e.g. #define MAX 1024). Third, protect your array bounds with the field-width modifier and don't put trailing whitespace in your format-string.
Putting it altogether (presuming you use 1024 as MAX, you can use
fscanf(fpt, "1023%s", temp))
Well done on checking the return of fscanf during your read.

Answer 2

Adding to the things that have already been mentioned.
In createWordCounter(...)
pCounter = (struct words*)malloc(sizeof(char));
you are allocating memory for a char. Even though the pointer to a struct is the pointer to its first member, the first element of words is a pointer to a char. It is better to be careful and write
struct words *pCounter = malloc(sizeof *pCounter);
Also, be mindful of operator precedence.
In addWord(...) you have
++pCounter->index;
What that does is increment the pointer pCounter before accessing index. If you are trying to increment index, it should be
++(pCounter->index);
or
pCounter->index++;
I recommend striping your program down to its bare essentials and test each part one at a time systematically to narrow down the cause of your errors.

Answer 3

I think the main problem is the size of temp array when you try to using fscanf.
while((fscanf(fpt, "%s ", temp)) == 1)
When the length of one line is bigger than MAX, segmentation fault occur.
You can change your code like this
#define SCANF_LEN2(x) #x
#define SCANF_LEN(x) SCANF_LEN2(x)
//...
//your original code
//...
while((fscanf(fpt, "%"SCANF_LEN(MAX)"s ", temp)) == 1)
By the way, you should check
(1) compile warning about type
char* removePunc(struct words* ch)
should be char* removePunc(char *ch)
if(temp == ' ') should be if(temp[0] == ' ')
if(temp == '\n') should be if(temp[0] == '\n')
(2) malloc size
pCounter = (struct words*)malloc(sizeof(char)); should be pCounter = (struct words*)malloc(sizeof(struct words));
(3) remember free after using malloc