Validation design

**AndiPersti** · 11-20-2012

Originally Posted by Salem

However, it seems to be almost completely useless.

It only seems to generate ERANGE if the number input exceeds the limits of long long int, regardless of the width of the datatype being converted at the moment.

Interesting. When I run your code I get:

Code:

$ ./foo
2147483647
9223372036854775807
Enter number: 2147483648
Number out of range, try again: Numerical result out of range
Enter number: 9223372036854775807
Number out of range, try again: Numerical result out of range
$ /lib/i386-linux-gnu/libc.so.6
GNU C Library (Ubuntu EGLIBC 2.13-20ubuntu5.2) stable release version 2.13, by Roland McGrath et al.

Originally Posted by Salem

If you have to read your data into the longest data type, then manually check the range to see if will fit in a smaller int, then you may as well use strtol() from the outset. It's the same PITA either way.

My first thought was using strtol() but because it's returning a long I took another look at the sscanf() man page to check if it does any error reporting and that's how I found out about the ERANGE.

Bye, Andreas

**AndiPersti** · 11-20-2012

Salem, are you on a 64bit system?

Bug 6453 – scanf integer overflow handling different between 32bit and 64bit

Bye, Andreas

**Salem** · 11-20-2012

Ah, I wonder if it's down to machine word size.

Code:

$ /lib/x86_64-linux-gnu/libc.so.6
GNU C Library (Ubuntu EGLIBC 2.13-20ubuntu5.2) stable release version 2.13, by Roland McGrath et al.

Here is something out of the glibc vfscanf,c file

Code:

	  /* Convert the number.  */
	  ADDW (L_('\0'));
	  if (need_longlong && (flags & LONGDBL))
	    {
	      if (flags & NUMBER_SIGNED)
		num.q = __strtoll_internal (wp, &tw, base, flags & GROUP);
	      else
		num.uq = __strtoull_internal (wp, &tw, base, flags & GROUP);
	    }
	  else
	    {
	      if (flags & NUMBER_SIGNED)
		num.l = __strtol_internal (wp, &tw, base, flags & GROUP);
	      else
		num.ul = __strtoul_internal (wp, &tw, base, flags & GROUP);
	    }

It seems to use the widest available strto... function to perform the initial conversion.

What follows next are just a bunch of dumb casts to throw away the msb data, ignoring the actual data type ranges.
Things like this:

Code:

		    *ARG (short int *) = (short int) num.l;

**Salem** · 11-20-2012

Originally Posted by AndiPersti

Salem, are you on a 64bit system?

Bug 6453 – scanf integer overflow handling different between 32bit and 64bit

Bye, Andreas

Mmm, "Status: RESOLVED INVALID", from 2008
In other words, not a bug, and nothing to fix.

It might just squeak in within the letter of the law, but it's certainly unhelpful with respect to the intent of the manual page enhancement.

Thread: Validation design

Thread Tools

Search Thread

Display

Similar Threads

Validation

validation in c

validation

need help in validation

Help with validation