C Program Help: Count & Sort Word Occurrences in File

[KV305]

C program help!

Hey guys I am having trouble completing this C program. Its suppose to ask the user to enter a file to read from, read the file, count how many times each word appears and print the results in a different file(also user input) arranged from high occurrence.


Heres what i have so far

#include <stdio.h>
#include <string.h>
#include <stdlib.h>

typedef struct AWORD
{
char wordname[50];
int count;
}a_word;

a_word nword(char* words)
{
a_word new;
strcpy (new.wordname, words);
return new;
}

int compare (words.wordname, words[j].wordname)
{
if (words.count < words[j].count)
{
return -1;
}
else if (words.count > words[j].count)
{
return 1;
}
else
{
return 0;
}
}

int main(int argc, char *argv[])
{
int i = 0,j = 0,k = 0, m = 0, count = 0,count2 = 0, count3 = 0, wrdcnt = 1,$
char cur, cur2[50], file1[25], file2[25], list[1000][k], *buffer;
FILE *pFile, *pFile2;

printf("Please type the name of the input file:\n ");
scanf("%s",file1);
printf("%s",file1);
pFile = fopen(file1, "r");

printf("Please type the name of the output file:\n ");
scanf("%s",file2);
pFile2 = fopen(file2, "r");

while (cur != EOF)
{
cur = fgetc(pFile);
if (cur != ' ')
{
list[m][k] = cur;
fprintf(pFile2, "%c", list[m][k]);
k++;
if (cur != '\n' && cur != '\0')
{
count2++;
}
if (cur == '\n' || cur == '\0')
{
wrdcnt++;
}
}
else
{
if (cur == ' ')
{
wrdcnt++;
}
m++;
k = 0;
if (cur == '\0')
{
fprintf(pFile2,"%s", &cur);
}
else
{
fprintf(pFile2, "\n");
}
}
}
fclose(pFile);
fclose(pFile2);

pFile2 = fopen(file2, "r");

a_word words[wrdcnt];

for (i = 0; i < wrdcnt; i++)
{
fgets (cur2, 50, pFile2);
words = nword(cur2);
printf("words[%d] is %s", i, words.wordname);
}
printf("\n");
count3 = wrdcnt;
for (j = 0; j < count3; j++)
{
buffer = words[j].wordname;
for (i = 0; i < wrdcnt; i++)
{
if (strcmp(buffer, words.wordname) == 0)
{
count++;
}
}

words[j].count = count;
count = 0;
}
printf ("\nWord Counter Program Starting...");
printf("\nWord Counter Program Finished. Check %s file\n", file2);

for (j = 0; j < count3; j++)
{
for (i = 0; i < j; i++)
{
if (strcmp(words[j].wordname, words.wordname) == 0)
{
same = 0;
break;
}
else
{
same = 1;
}
}
if (same == 1)
{
printf ("%d\t%s", words[j].count, words[j].wordname);
fprintf(pFile2, "%d %s\n", words[j].count, words[j].wordname);

}

}
fclose(pFile2);
return 0;
}




I haven't sorted yet. we are suppose to use qsort but i don't think my compare function is right. Any help would be greatly appreciated. Thank you.

 

[AJ Bentley]

You have to create a list of words and a number for each as you have done (AWORD).

You then need to do a bubble sort - going through your list of words swapping over any pair that are the wrong way round - you do that multiple times until there are no swaps - then you know it's done.
So you need to compare a pair of integers - the number of times each word occurs in each pair.

The declaration for the compare must use a void *. (it can't cope with overloading)
You will be passing it a pair of integers so you need to cast each to a const void* in the qsort call.
Then your declaration will look like:-

int compare (const void *count1, const void *count2)
{
//then you cast it back to the type you want - you can do that on-the-fly
// I think you can see where it goes from here
if ((int)count1==(int)count2) return 0; //etc.
}

 

[KV305]

Ok so far i altered the program and gives no errors but when i try to run it it says
"Segmentation fault"... any ideas?

#include <stdio.h>
#include <string.h>
#include <stdlib.h>

typedef struct AWORD
{
char wordname[50];
int count;
}a_word;

a_word nword(char* words)
{
a_word new;
strcpy (new.wordname, words);
return new;
}

int compare (const void *count1, const void *count2)
{
if ((int)count1==(int)count2) return 0;
if ((int)count1<(int)count2) return -1;
if ((int)count1>(int)count2) return 1;
}

int main(int argc, char *argv[])
{
int i = 0,j = 0,k = 0, m = 0, count = 0,count2 = 0, count3 = 0, wrdcnt = 1,same;
char cur, cur2[50], file1[25], file2[25], list[1000][30], *buffer, sorted[1000];
FILE *pFile, *pFile2;

printf("Please type the name of the input file:\n ");
scanf("%s",file1);
printf("%s",file1);
pFile = fopen(file1, "r");

printf("Please type the name of the output file:\n ");
scanf("%s",file2);
pFile2 = fopen(file2, "r");

while (cur != EOF)
{
cur = fgetc(pFile);
if (cur != ' ')
{
list[m][k] = cur;
fprintf(pFile2, "%c", list[m][k]);
k++;
if (cur != '\n' && cur != '\0')
{
count2++;
}
if (cur == '\n' || cur == '\0')
{
wrdcnt++;
}
}
else
{
if (cur == ' ')
{
wrdcnt++;
}
m++;
k = 0;
if (cur == '\0')
{
fprintf(pFile2,"%s", &cur);
}
else
{
fprintf(pFile2, "\n");
}
}
}
fclose(pFile);
fclose(pFile2);

pFile2 = fopen(file2, "r");

a_word words[wrdcnt];

for (i = 0; i < wrdcnt; i++)
{
fgets (cur2, 50, pFile2);
words = nword(cur2);
printf("words[%d] is %s", i, words.wordname);
}
printf("\n");
count3 = wrdcnt;
for (j = 0; j < count3; j++)
{
buffer = words[j].wordname;
for (i = 0; i < wrdcnt; i++)
{
if (strcmp(buffer, words.wordname) == 0)
{
count++;
}
}

words[j].count = count;
count = 0;
}
printf ("\nWord Counter Program Starting...");
printf("\nWord Counter Program Finished. Check %s file\n", file2);

for (j = 0; j < count3; j++)
{
for (i = 0; i < j; i++)
{
if (strcmp(words[j].wordname, words.wordname) == 0)
{
same = 0;
break;
}
else
{
same = 1;
}
}
if (same == 1)
{
printf ("%d\t%s", words[j].count, words[j].wordname);
fprintf(pFile2, "%d %s\n", words[j].count, words[j].wordname);
}

}
fclose(pFile2);
pFile2 = fopen(file2, "r");
m=0;
while (pFile != NULL)
{
fscanf (pFile2 , "%s" , &sorted[m]);
m++;
}
qsort(sorted,m,sizeof(int),compare);
fclose(pFile2);
pFile2 = fopen(file2, "w");
fprintf(pFile2,"%s", sorted);
return 0;
}

 

[Mark44]

Are you in the same class as the person who started this thread?

If you're getting a segmentation fault, you are probably attempting to access memory (via a pointer) that your program is not allowed to access. It's time to learn how to use a debugger.

 

[jtbell]

when i try to run it it says
"Segmentation fault"... any ideas?

Put extra printf() statements in the program so you can pin down exactly where the segmentation fault is occurring:

"I got to checkpoint #1.
I got to checkpoint #2.
I got to checkpoint #3.
ERROR: Segmentation fault."

 

[AJ Bentley]

It means you are attempting to read or write to a part of the memory outside your own program. Usually a pointer error (accessing memory location zero because a pointer is null for example).

Try temporarily commenting out sections of your program till you locate the line giving the error - then look carefully at any pointers involved.

Doesn't your debugger automatically stop at error points?