OpenEuphoria

Re: Round

• Posted by Aidan Bindoff <abindoff at ONE.NET.AU> Jun 10, 2000
• 609 views

Lucius, thankyou for pointing that out. In my applications I hadn't even
considered atoms that big. Also, making a loop does seem pretty convoluted
when you can just use force_int(object) - but that's Euphoria!  You don't
always wake up and smell the coffee with everything different thing you
do...

Kind Regards,
Aidan
----- Original Message -----
From: Lucius L. Hilley III <lhilley at CDC.NET>
To: <EUPHORIA at LISTSERV.MUOHIO.EDU>
Sent: Friday, June 09, 2000 10:55 PM
Subject: Re: Round


>     I have some things to address here that I am surprised haven't been
> mentioned by someone else yet.
>
> Your code....
>     sqrt(number*number)--abs()
> I only see one flaw in this absolute value code.  Yes, it works.  Yes, it
> will most likely work in every case that you ever use it in.  No it
doesn't
> work for all atoms.  In fact it doesn't come close to working with all
> atoms.
>
> Example:
> I start with a with a value that is much less than half the atoms
possible.
>
>   a = largest_possible_atom / 3
>   a = sqrt(a*a)--Oops....  the poor thang crashed.
>
> --lets say way break the above line down and see why.
>   a = (a*a)--Oops... the poor than crashed...
>
> --Now let's analyze why.
> TRUE or FALSE !!!
>   largest_possible_atom is greater than 3?
>   (6/2) * (6/2) equals 9?
>   9 is greater than 6?
> Assume that 6 is the largest_possible_atom
> You know that 9 is greater than 6.
> Therefore 9 is large enough to crash the program.
>
> You know that Half of 6 is 3 AND
> 3 times 3 = 9
> You can see that your math will keep more than half of the possible atoms
> from being evaluated.
>
> function your_abs(atom a)--You
>   return sqrt(a * a)
> end function
>
> function my_abs()
>   if a > 0 then
>     return a
>   else
>     return -a
>   end if
> end function
>
>
>
> NEXT
>
> You said "I also have this in a loop to round strings".  I assume you mean
> the following example.
>
> i =  force_integer(4.3) -- i = 4
> s = repeat(i, 3)--{4.3, 4.3, 4.3}
> for A = 1 to length(s) do
>   s[A] = force_integer(s[A])
> end for
> --s = {4, 4, 4}
>
> If this is the case and the method in which you handle a long set of
> integers, then wake up and realize the power of sequences.  The math will
> work on an entire sequence just as easily as it will work on and
individual
> atom.
>
> In your code listed below just change:
>
> function force_integer(atom number)
>      TO
> function force_integer(object number)
>
>         Lucius L. Hilley III
>         lhilley at cdc.net
> +----------+--------------+--------------+
> | Hollow   | ICQ: 9638898 | AIM: LLHIII  |
> |  Horse   +--------------+--------------+
> | Software | http://www.cdc.net/~lhilley |
> +----------+-----------------------------+