This is an odd situation.
It works properly on a D3 system (which doesn't prove anything).
I tried it on UniVerse release 10.0.10 / NT, and it fails as reported.
I treid it on UniVerse release 9.6.4.1 on RH 7 and it fails as reported
What I did realize after running this a few times (although this didn't make
sense either),
is that if you change:
DEC.OFFSET = 10000
to: DEC.OFFSET = 1000, It works properly
The numbers match to the eye in all cases.
If there was a .0000000001 difference, I would say that there is a
difference, and the result should fail, but the values match precisely to
the eye.
One should NOT have to add a zero to make the numbers match.
I vote it's a bug.
Lee Bacall
http://www.binarystar.com
Phone: +1 (954) 791-8575
Cell: +1 (954) 655-6581
> -----Original Message-----
> From: Chris van O [mailto:[EMAIL PROTECTED]
> Sent: 08 October 2004 07:30
> To: [EMAIL PROTECTED]
> Subject: [U2] Why is this statement failing? (UniVerse)
>
>
> Ive found a bug that has been in production for many years, but has just
> started to cause problems....but I dont know exactly why...
>
> DEC.OFFSET = 10000
> FOR I = 1 TO 20
> FOR J = 1 TO 99
> K = I "R0%2" : '.' : J "R0%2"
> PRX = K
> IF INT(PRX*DEC.OFFSET)=DEC.OFFSET*PRX THEN
> PRINT K : ' PASSED'
> END ELSE
> PRINT K : ' FAILED'
> END
> NEXT J
> NEXT I
>
> Ive fixed it by saying .....
>
> IF (PRX*DEC.OFFSET)+0 = DEC.OFFSET*PRX THEN
-------
u2-users mailing list
[EMAIL PROTECTED]
To unsubscribe please visit http://listserver.u2ug.org/
