I just discovered that RC4 for Windows does not issue an
error message if one tries to access array element nr. -1

Here's the code I tried:


#declare A = array[1] { 0 }

// No error messages
#declare N = A[-1]; 
#declare A[-1] = 0;

// "Parse Error: Negative subscript"
#declare N = A[-2]; 
#declare A[-2] = 0;


If this is not old news, then can anyone please confirm this ?


Tor Olav

Post a reply to this message

On Thu, 09 May 2002 01:55:58 +0200, Tor Olav Kristensen
<tor### [at] hotmailcom> wrote:
> I just discovered that RC4 for Windows does not issue an
> error message if one tries to access array element nr. -1
>
> Here's the code I tried:
>
> #declare A = array[1] { 0 }
>
> // No error messages
> #declare N = A[-1]; 
> #declare A[-1] = 0;

I can confirm however I wonder if it could be ... floating point accuracy.
What rounding is performed when float is converted to integer ? Truncation
towards zero ? Could it be that -1 is internally something like -0.99999999999
and after truncation it is valid value 0 ? If yes then possible fix is to
check sign of the float before conversion to int.

POV 3.5 RC4 icl on WinNT Sp 6 PII 233 with 128 MB

ABX

Post a reply to this message

Skiba <abx### [at] babilonorg>  wrote:

> I can confirm however I wonder if it could be ... floating point accuracy.
> What rounding is performed when float is converted to integer ? Truncation
> towards zero ? Could it be that -1 is internally something like -0.99999999999
> and after truncation it is valid value 0 ?

The rounding method used requires a small bias to be added.  That is added
always.  It can of coure be fixed.

    Thorsten

____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trfde

Visit POV-Ray on the web: http://mac.povray.org

Post a reply to this message