snprintf()

Lightweight and with minimal dependencies implementation of snprintf() C function. Especially it is independend from mathematical functions from math.h which are not always available for embedded platforms.

From many years I was using almost unchanged orginal implementation of Alain Magloire (v1.1) but last time I needed replacement of snprintf() function more compliant with standard implementation.

Function prototype

int snprintf(char *string, size_t length, const char *format, ...);

Input parameters

Parameter	Description
`string`	Output buffer.
`length`	Size of output buffer `string`.
`format`	Format of input parameters.
`...`	Input parameters according to `format`.

Return values

Value	Description
>=0	Amount of characters put (or would be put in case of `string` is set to `NULL`) in `string`.
-1	Output buffer `string` size is too small.

How to use it

Just copy src/snprintf.c & include/snprintf.h to your project and include snprintf.h in your code.

#include <stdio.h>

#include "snprintf.h"
  
int main(int argc, char *argv[]) {
  const char *hello = "Hello", *world = "World";
  char msg[0x100] = "";

  snprintf(msg, sizeof(msg), "%s %s!", hello, world);
  printf("%s\n", msg);
  
  return 0;
}

In src/tests-snprintf.c & include/tests-snprintf.h you can find set of tests based on MinUnit engine, which can be run by tests_snprintf() function.

#include "tests-snprintf.h"

int main(void) {
  return tests_snprintf();
}

Supported format specifiers

Supportted types

Type	Description
d / i	signed decimal integer
u	unsigned decimal integer
o	unsigned octal integer
x / X	unsigned hexadecimal integer
f / F	decimal floating point
e / E	scientific (exponential) floating point
g / G	scientific or decimal floating point
c	character
s	string
p	pointer
%	percent character

Supported lengths

Length	Description
hh	signed / unsigned char
h	signed / unsigned short
l	signed / unsigned long
ll	signed / unsigned long long

Supported flags

Flag	Description
-	justify left
+	justify right or put a plus if number
#	prefix 0x, 0X for hex and 0 for octal
*	width and/or precision is specified as an int argument
0	for number padding with zeros instead of spaces
(space)	leave a blank for number with no sign

Authors

Mirosław Toton, mirtoto@gmail.com
Alain Magloire, alainm@rcsm.ee.mcgill.ca

mirtoto/snprintf