begin  quoting Gabriel Sechan as of Wed, Jun 07, 2006 at 08:48:13PM -0500:
[snip]
> He's wrong about the results of the bug-  he says in C it overflows by 
> going to an invalid index via underflow.  In C, you'd use an unsigned int 
> (does Java have this?)

No, Java does not have an unsigned int.  But then, there's no problem
with using a negative index into an array -- you just get an exception.

When someone finally *does* demonstrate this, it'll be a quick
identification of what went wrong and where.  I love Java stack traces.
(Fought with a C++ app today that had a permission problem when reading
a file, so I got a cryptic "Program failed to initialize" error... Gah!)

>                        thus wouldn't go to a negative index (although it is 
> still a bug, it won't crash the app.  It may infinite loop).

Heh. Been there, done that, back in one of my first C programs, IIRC.

>                                                               IF a variable 
> is supposed to be a loop index, it should always be unsigned in any 
> language.

Erm, no.

Even for array indexes, there are languages where that's allowed.

There's a disconnect between the hardware and the software and our
minds. WE don't put an upper limit on numbers. Hardware does. Software
is in between. . .

-- 
_ |\_
 \|

-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg

Reply via email to