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

#define NULLCHK(S) if ((S) != NULL)\
exit(1)

/*
* A basic implementation of seq(1). Supports neither floating-point bounds nor
* custom increment values.
*/
int
main(int argc, char **argv) {
if (argc < 2 || argc > 3)
exit(2);

long long lo, hi;
const char *errstr = NULL;

lo = (argc == 2) ? 1 : strtonum(argv[1], LLONG_MIN, LLONG_MAX, &errstr);
NULLCHK(errstr);

hi = strtonum(argv[argc - 1], LLONG_MIN, LLONG_MAX, &errstr);
NULLCHK(errstr);

for (long long i = lo; i <= hi; i++)
printf("%lld\n", i);

exit(0);
}

static void
incr (char **s0, size_t *s_len)
{
char *s = *s0;
char *endp = s + *s_len - 1;

do
{
if ((*endp)++ < '9')
return;
*endp-- = '0';
}
while (endp >= s);
*--(*s0) = '1';
++*s_len;
}

@mallaidh why

@KitRedgrave because gnu utils tend to be horrifyingly overcomplicated as well as being originally written to avoid accusations of copying then-proprietary unix sources

compare the man pages:
http://man.cat-v.org/unix_8th/1/seq
https://linux.die.net/man/1/seq

i decided to make the whole thing infinitely more simple by dropping floating-point entirely, because i'll only be using it as a for-loop replacement thing in ksh scripts