C-如何将文本文件中的单词读入字符串数组

Question

I need to write a program that generates a table mapping words onto the number of times the word appears in a text file. So far my code looks like this

#include <stdlib.h>
#include <ctype.h>
#include <stdio.h>
#include <string.h>
struct entry
{
  char* word;
  unsigned int n;
  struct entry *left;
  struct entry *right;
};

struct entry* 
insert(struct entry *table, char *str)
{
  if(table==NULL){
    table = (struct entry*)malloc(sizeof(struct entry));
    table->word = str;
    table->n = 1;
    table->left = NULL;
    table->right = NULL;
  }else if(strcmp(table->word,str)==0){
    table->n=(table->n)+1;
  }else if(strcmp(table->word,str)==1){
    table->left=insert(table->left,str);
  }else{
    table->right = insert(table->right,str);
  }
  return table;
}

void
print_table(struct entry *table)
{
  if(!(table==NULL)){
    print_table(table->left);
    fprintf(stdout,"%s\t %d\n",table->word,table->n);
    print_table(table->right);
  }
}

int
main(int argc, const char *argv[])
{

  struct entry* table = NULL;
  char *str = "foo";
  table =  insert(table,str);
  str = "foo";
  table = insert(table,str); 
  print_table(table);

  return 0;

}

which gives an out put of

foo 2

what I need to do is do this exact thing with an input file. My idea is to take every word of the text file which will look like

This is an example of 
what the text file
will look like.

I have no idea what the exact number of lines or words per line are. As I was saying, my idea was to take every word from the text file and put it into an array of strings, then run my insert function through every element in the array, I just have no idea how I should go about taking each word and putting it in the array. Any suggestions are welcome and appreciated.

Answer 1

If you want to store every word of the following paragraph

This is an example of 
what the text file
will look like.

The following will work:

while(true){
    while(inFile >> yourword){
    //store yourword here
    }
    getline(inFile, yourword); //discards the newline
    if(/*some_conditional_to_break*/)
    break;
}

Answer 2

Portability bug

Note that this use of strcmp() is wrong:

}else if(strcmp(table->word,str)==1){

The definition of strcmp() is that it returns a value less than zero, equal to zero, or greater than zero. No mention of 1 .

Always, but always, compare with 0:

• if (strcmp(word, str) == 0) — word equal to str
• if (strcmp(word, str) != 0) — word not equal to str
• if (strcmp(word, str) <= 0) — word less than or equal to str
• if (strcmp(word, str) >= 0) — word greater than or equal to str
• if (strcmp(word, str) < 0) — word less than str
• if (strcmp(word, str) > 0) — word greater than str

In many implementations, the return value from strcmp() is the numeric difference between the characters that differ, and can be much larger or smaller than 1.

Reading words

If you're reasonably sure your input won't be wholly insane, you can use a variant on this loop to read the data:

char buffer[4096];
while (fscanf(fp, "%4095s", buffer) == 1)
{
    char *word = strdup(buffer);
    table = insert(table, word);
}

This reads words up to 4 KiB long and stores each one in your table using your function. If a word is exactly 4 KiB long or longer than that, it will be split into pieces. It probably won't be a problem. Note that the scanf() family treats blanks, tabs and newlines as the separators between words. Writing "az" in a file gets treated as one word, double quotes, dash and all.