entab.c - Program that replaces a string of blanks by the min # of spaces & tabs required to generate identical output

entab.c

// http://forum.sa-mp.com/showthread.php?t=649767
#include <stdio.h>                                                                                                                                                       

#define TABSIZE 8

int main()
{
    int c, cft, nspaces;

    //cft = chars from tab, nspaces = number of spaces buffered
    cft = nspaces = 0;
    while((c = getchar()) != EOF) {

        if(c == '\n') {
            // reset ALL the things
            cft = nspaces = 0;
            putchar(c);
            continue;
        }

        if(c == '\t') {
            // treat tabs as spaces to easily minify later
            // don't add TABSIZE, calculate the amount of chars to the next tab column
            nspaces += TABSIZE - (cft + nspaces) % TABSIZE;
            continue;
        }

        if(c == ' ') {
            ++nspaces;
            continue;
        }

        //decide whether gap in between chars is long enough for a tab
        while(nspaces >= TABSIZE - cft) {
            putchar('\t');
            nspaces -= TABSIZE - cft;
            cft = 0;
        }

        //insert all remaining spaces manually
        while(nspaces > 0) {
            putchar(' ');
            --nspaces;
            ++cft;
        }

        cft = (cft + 1) % TABSIZE;

        putchar(c);
    }

    return 0;
}

Hey! Thanks for the help man! After some discussion on CodeReview on StackExchange, I got some feedback with which I've made the 2nd revision.

A couple of comments just to see if we're on the same track:

• After while(nspaces > 0, since we're decrementing it in steps of one, we're certain that nspaces must be 0.
• I like how you are using the % operand to calculate tabsizes, but the line nchars = 0; works to fix the logic error where any line with TABSIZE (8) + nonwhitespace characters would be given extra tabs.
• I am glad you showed me your version using continue, but the approach in the 2nd version makes more sense to me than this one.

Thanks a lot nonetheless, I am very happy you spent time in helping me :)

• nspaces can be negative because you check for nspaces >= TABSIZE - nchars but subtract the full TABSIZE Oh I guess you're talking about my code? yeah I removed that statement when I saw that later
• the nchars = 0 doesn't fix everything I think

I discussed it more thoroughly in my reply
http://forum.sa-mp.com/showthread.php?p=3993467#post3993467