Hi,
On Mon, 2002-11-11 at 04:05, Randy.Dunlap wrote:
> On Sun, 10 Nov 2002, Henning P. Schmiedehausen wrote:
>
> | "Randy.Dunlap" <rddunlap@osdl.org> writes:
> |
> | >+ digit = *str;
> | >+ if (is_sign && digit == '-')
> | >+ digit = *(str + 1);
> |
> | If signed is not allowed and you get a "-", you're in an error case
> | again...
>
> Yes, and a 0 value is returned.
> IOW, asking for an unsigned number (in the format string)
> and getting "-123" does return 0.
>
> What should it do?
I would model this after user space. (Which does strange things:
--- cut ---
#include <stdio.h>
main()
{
char *scan = "-100";
unsigned int foo;
int bar;
int res = sscanf(scan, "%ud", &foo);
printf("%s = %ud = %d\n", scan, foo, res);
res = sscanf(scan, "%ud", &bar);
printf("%s = %d = %d\n", scan, bar, res);
}
--- cut ---
% gcc -o xxx xxx.c
./xxx
-100 = 4294967196d = 1
-100 = -100 = 1
Hm, so I guess, yes, a warning message would be nice IMHO. Returning an
error code would IMHO moot, because noone is checking these codes
anyway.
Regards
Henning
> This function can't return -EINPUTERROR or -EILSEQ.
> (since it's after feature-freeze :)
> And the original problem was that a leading '-' sign on a
> signed number (!) caused a return of 0. At least that is fixed.
>
> So now the problem (?) is that a '-' sign on an unsigned number
> returns 0. We can always add a big printk() there that
> something is foul. Other ideas?
-- Dipl.-Inf. (Univ.) Henning P. Schmiedehausen -- Geschaeftsfuehrer INTERMETA - Gesellschaft fuer Mehrwertdienste mbH hps@intermeta.deAm Schwabachgrund 22 Fon.: 09131 / 50654-0 info@intermeta.de D-91054 Buckenhof Fax.: 09131 / 50654-20
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Nov 15 2002 - 22:00:21 EST