Print a Histogram based on word lengths (C)

Question

This is a K&R exercise (1-13)...

"Write a program to print a histogram of the length of words in its input. It is easy to draw the histogram with bars horizontal; a vertical orientation is more challenging."

The section was about arrays, and to be honest, I'm not sure I fully understood it. Everything up to this point was fairly easy to grasp, this was not.

Anyway I'm trying to do a histogram with horizontal bars first. Once I got that down I'll try vertical, but right now I'm not even sure where to begin with the easy version. (I slept on it, woke up, and still couldn't get it.)

I drew an example of what the program would output:

----------------------------------------------------------------
001|XX
002|XXXX
003|X
004|XXXXXXXXXX
005|XXXXXXXXXXXXXXXXXXXXXXXXX
006|XXXX
007|X
008|
009|XXXXXXXXX
010|XXX
>10|XXXX
----------------------------------------------------------------

And tried to break it (the program) down in sections. This is what I came up with:

1. PRINT TOP BORDER
2. PRINT CATEGORY, PRINT X EACH TIME CONDITION IS TRUE, PRINT NEWLINE, REPEAT.
3. PRINT BOTTOM BORDER

But the more I think about it the less I think that's how it would work (because getchar() goes through one character at a time, and it wouldn't be able to go back up to put a X in the right category.) Or...

... I'm just really confused as to how I would solve this problem. Here's as far as I've been able to get code wise:

#include <stdio.h>

#define MAXWORDLENGTH 10

// print a histogram of the length of words in input. horizontal bar version

int main(void)
{
  int c;
  while ((c = getchar()) != EOF) {

  }

  return 0;
}

Could someone help enlighten me? Not necessarily with the code, maybe just pseudo code, or with some "words from the wise" as to what I need to do, or think, or something. This has just been a really big stone in the road and I'd like to get past it :/.

(I'll check back in 30 minutes)

Answer 1

I loved the pseudo-code! Some good thinking there, but you're still not ordering your program right.

As you said yourself, you can't read the text, go back and print an X in a particular row. If we establish that it can't be done, then there's no choice but to know all the values of the histogram beforehand.

So you should think your program as having two parts (and you'll make this kind of division in practically every program you write): first, a part that will make calculations; and then a part that will output them in a certain format (the histogram).

This tip should get you started! If you need further help, comment below.

Answer 2

I suggest you simplify the problem by solving it for the case of one word per line, so you can use fgets. Here's how to "eat up" lines that are too long.

Then, as often, the central data structure is the key to solving the problem. The data structure you need is an array used as frequency table:

int freq[11];

In freq[1], store the number of words/lines of length 1, in freq[2] those of length 2, etc., and in freq[0] those of length >10. You don't need to store the words since the rest of the program only needs their length. Writing out the histogram should be easy now.

I hope this isn't too much of a spoiler.