Help with reading a file

[kgos]

Iam trying to read a file, count the number of words in it and list each word with it's frequency. I used the code bellow but when l try to run it, it just hungs and nothing is outputed. lam relatively new in c/c++ field trying to do it by way of self study. I will appliciate the help


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

#define MAXWORD 10000


struct tnode {
char *word;
int count;
struct tnode *left;
struct tnode *right;
};

struct tnode *addtree(struct tnode *, char *);
void treeprint(struct tnode *);
struct tnode *talloc(void);
int getword(char *, int);
char *strdupl(char *);


/* word frequency count */

int main(void)
{
struct tnode *root;
char word[MAXWORD];
root = NULL;

while (getword(word, MAXWORD) != EOF)
if (isprint(word[0]))
root = addtree(root, word);
treeprint(root);
exit (0);
}

/* addtree: add a node with w, at or below p */

struct tnode *addtree(struct tnode *p, char *w)
{
int cond;

if (p == NULL) { /* a new word has arrived */
p = talloc(); /* make a new node */
p->word = strdupl(w);
p->count = 1;
p->left = p->right = NULL;
}
else if ((cond = strcmp(w, p->word)) == 0)
(p->count)++; /* repeated word */
else if (cond < 0) /* less than into left subtree */
p->left = addtree(p->left,w);
else /* greater than into right subtree */
p->right = addtree(p->right,w);
return p;
}

/* treeprint: in-order print of tree p */

void treeprint(struct tnode *p)
{
if (p !=NULL) {
treeprint(p->left);
printf("%4d %s\n", p->count, p->word);
treeprint(p->right);
}
}


/* talloc: make a tnode */

struct tnode *talloc(void)
{
return (struct tnode *) malloc(sizeof(struct tnode));
}


/* getword: get next word or character from input */

int getword(char *word, int lim)
{

FILE *pFile;
int c;
char *w = word;

pFile = fopen ("/mnt/samba/k/patrick/testp.txt","r");


while (!isspace(c = fgetc(pFile)) && c!=EOF)


*w++ = tolower(c);

if (!isprint(c))
{
*w = '\0';
return c;
}
for ( ; --lim > 0; w++)
if(!isalnum(*w = fgetc(pFile)) && !ispunct(*w))
{
ungetc(*w,pFile);
break;
}

*w = '\0';
fclose(pFile);
return word[0];
}

/* strdupl: make duplicate of s. (strdup builtin) */

char *strdupl(char *s)
{
char *p;
p = (char *) malloc(strlen(s)+1);
if (p != NULL)
strcpy(p, s);
return p;
}

[Magos]

It seems like you're reopening the file in every call to GetWord().
Oh, and use code tags when posting code (read the FAQ).

You could try this pseudo code:

OpenFile();

while(Looping)
{
   GetWordFromFile();

   if(FindWordInList())
   {
      AddOneToCounterOfThatWord();
   }
   else
   {
      AddWordToList();
   }

   if(EndOfFile()) Looping = false;
}

CloseFile();

[Magos]

BTW, I don't think so many words contain 10000 letters .

[Prelude]

Change the prototype of getword to this:

int getword(char *, int, FILE *);

The definition to this:

int getword(char *word, int lim, FILE *pFile)
{
  int c;
  char *w = word;
  
  while (isspace(c = fgetc(pFile)))
    ;
  
  if (c != EOF)
    *w++ = c;
  
  if (!isalpha(c))
  {
    *w = '\0';
    return c;
  }
  
  for ( ; --lim > 0; w++)
  {
    if(!isalnum(*w = fgetc(pFile)))
    {
      ungetc(*w, pFile);
      break;
    }

    *w = tolower(*w);
  }
  
  *w = '\0';
  
  return word[0];
}

And main to this:

int main(void)
{
  FILE *pFile;
  struct tnode *root;
  char word[MAXWORD];
  
  root = NULL;
  pFile = fopen("/mnt/samba/k/patrick/testp.txt", "r");
  
  if (pFile != NULL)
  {
    while (getword(word, MAXWORD, pFile) != EOF) {
      if (isalpha(word[0]))
        root = addtree(root, word);
    }

    fclose(pFile);
  }
  else
    perror("File open failure");
  
  treeprint(root);
  exit (0);
}

You chose a good book to learn from, K&R is the bible for C.

-Prelude