On AVR 8-bit, avoiding floats, I'm dividing a signed integer and formatting it with snprintf:

int16_t tempx10 = -5;
div_t temp = div(tempx10, 10);
char buf[10];
snprintf(buf, sizeof (buf), "%4d.%d°", temp.quot, abs(temp.rem));

This however produces 0.5 instead of -0.5.

What would be an elegant way to solve this?

On AVR 8-bit, avoiding floats, I'm dividing a signed integer and formatting it with snprintf:

int16_t tempx10 = -5;
div_t temp = div(tempx10, 10);
char buf[10];
snprintf(buf, sizeof (buf), "%4d.%d°", temp.quot, abs(temp.rem));

This however produces 0.5 instead of -0.5.

What would be an elegant way to solve this?

• printf
• avr-gcc

• 1 soo "%s...", tmpx10 < 0 ? "-" : "+", ....? – KamilCuk Commented Mar 7 at 11:44
• that produces something like - -1.5° – Torsten Römer Commented Mar 7 at 11:51
• there's no easy way to justify - with 1 to the right. You have to write code to handle it, yes. – KamilCuk Commented Mar 7 at 12:00

2 Answers 2

The code that you provided omitted the negative sign when temp.quot was zero because div() returns a zero quotient for -5/10, and the format string didn't account for the original value's sign. You can add the sign manually and you can use abs(temp.quot) to prevent the quotient from repeating its sign. You can do it this way.

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

int main() {
    int16_t tempx10 = -5;
    div_t temp = div(tempx10, 10);
    char buf[10];

    snprintf(buf, sizeof(buf), "%c%d.%d°",
             (tempx10 < 0) ? '-' : ' ',
             abs(temp.quot),    
             abs(temp.rem));

    printf("Result: %s\n", buf);  
    return 0;
}

Output:

Result: -0.5°

The code you are using is quite costly on AVR (division, sprintf), so here is a solution that only invokes some character operations. The printf is only used to print the result.

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

void prt (int16_t tempx10)
{
    char buf[10];
    itoa (tempx10, buf, 10);
    size_t len = strlen (buf);

    // Position of the last digit.  Will become digit after '.'.
    char *p = buf + len - 1;
    char decimal = *p;
    if (tempx10 < 10 && tempx10 > -10)
        *p++ = '0';
    *p++ = '.';
    *p++ = decimal;
    *p++ = '\0';

    printf ("Temp = %s\n", buf);
}

int main (void)
{
    for (int i = -15; i <= 15; i+= 5)
        prt (i);
    return 0;
}

Ouput

Temp = -1.5
Temp = -1.0
Temp = -0.5
Temp = 0.0
Temp = 0.5
Temp = 1.0
Temp = 1.5