Show some decimal places C

Question

I'm coding a simple program for class. I've completed it, don't worry, I'm not asking for anyone to do my homework. Let me explain what I want with an example.

My program asks for an amount of bits and converts it to mb, kb and bytes. So, the output if I input 1 bit is:

1 in megabytes is: 0.000000119209290
1 in kilobytes is: 0.000122070312500
1 in bytes is: 0.125000000000000
1 in bits is: 1

So, my question is just an aesthetic one: how could I not show the decimal places that are unnecessary? For example, in the bytes, I would like to only print 0.125, instead of 15 decimals, which isn't that pretty at all.

The source code is:

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

int main(void)
{

        unsigned long long int bits;

        printf("Input a quantity of bits: \n");
        scanf("%lld", &bits);

    /*
     * 1 byte = 8 bits.
     * 1 kilobyte = 1024 bytes.
     * 1 megabyte = 1024 kilobytes.
     */
        long double by = ((double) bits) / ((double) 8);
        long double kb = ((double) by)  / ((double) 1024);
        long double mb = ((double) kb) / ((double) 1024);

        printf("%lld in megabytes is: %.15Lf\n", bits, mb);
        printf("%lld in kilobytes is: %.15Lf\n", bits, kb);
        printf("%lld in bytes is: %.15Lf\n", bits, by);
        printf("%lld in bits is: %lld\n", bits, bits);

    return(0);
}

PS: I know I specified 15 decimal places in the printf, I wast just trying which was the best way for me to output the values.

Thank you in advance!

http://stackoverflow.com/questions/277772/avoid-trailing-zeroes-in-printf This may help you. — venki421, Sep 28 '14 at 13:53
THe thig is that using %g I get the megabytes as an exponential number, and I want a decimal one. — Ignasi Sánchez, Sep 28 '14 at 13:55
Did you read [printf(3)](http://man7.org/linux/man-pages/man3/printf.3.html) — Basile Starynkevitch, Sep 28 '14 at 13:56

score 4 · Answer 1 · edited May 23 '17 at 11:49

Use g specifier, like this:

printf("%lld in megabytes is: %.15Lg\n", bits, mb);
printf("%lld in kilobytes is: %.15Lg\n", bits, kb);
printf("%lld in bytes is: %.15Lg\n", bits, by);
printf("%lld in bits is: %lld\n", bits, bits);

However, this will use scientific notation if needed. You could add an if statement like this:

if(fmod(mb, 10) == mb // last digit is not zero
    && mb < 0.000001) // if mb is a small number (the left value may need tuning)
   printf("%lld in megabytes is: %.15Lf\n", bits, mb);
else
   printf("%lld in megabytes is: %.15Lg\n", bits, mb);

A relevant answer is this. Also notice, that I had to use fmod() (under math.h), because mb is not an integer.

You are welcomed @IgnasiSánchez. You also made a good question, thus my +1. :) — gsamaras, Sep 28 '14 at 18:24

score 1 · Accepted Answer · answered Sep 28 '14 at 14:42

You could make a function to calculate the required number of decimal places. To do it, take the fractional part, and multiply it by 10 until it becomes an integer.

int required_decimal_places(double x)
{
    int counter = 0;

    x -= floor(x);
    while (x != floor(x))
    {
        x *= 10;
        ++counter;
    }

    return counter;
}

Then, output your number with the required number of decimal places:

printf("%lld in megabytes is: %.*f\n", bits, required_decimal_places(mb), mb);

The asterisk (*) in the format string tells the system that the length of output is specified as a parameter.

Note: I replaced long double in your code by double because I am not sure whether it is correct to call the library function floor on long double. I also changed Lf to f inside the format string. The extra accuracy of long double (if any) is not required in the calculations you are doing here.

This one is pretty genious, and it's the one I finally used. Thank you very much! — Ignasi Sánchez, Sep 28 '14 at 17:17

Basile Starynkevitch · Answer 3 · 2014-09-28T14:05:46.353

A possible way might be to print into a string, then to check that the string is accurate enough:

double x= some_number();
char buf[48];
snprintf (buf, sizeof(buf), "%.3f", x);
if (atof(buf) == x) puts(buf);
else {
  snprintf (buf, sizeof(buf), "%.6f", x);
  if (atof(buf) == x) puts(buf);
  else printf("%.15g", x);
}

Read floating point guide; consider comparing with some epsilon e.g perhaps something like if (abs(x-atof(buf)) < 1.0e-5*abs(x))

BTW, be aware of bignum-s and consider GMPlib if you need a lot (more than 8) decimals.

Show some decimal places C

3 Answers3