• Previous message: Conrad H Ziesler: "magscale 1 -2 bug"
• Maybe in reply to: Conrad H Ziesler: "magscale 1 -2 bug"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Dear Conrad,

   The rigorous solution should be:

-----------------------------------------------------

int
FindGCF(a, b)
    int a, b;
{
    register int a_mod_b, bp;

    bp = abs(b);
    if ((a_mod_b = (abs(a)) % bp) == 0) return (bp);
    else return (FindGCF(bp, a_mod_b));
}

-----------------------------------------------------

Thanks for pointing out the bug.  I've never run into that problem before,
but I guess it comes from running the greatest-common-factor routine on
negative coordinates or something.

For anyone else tuning in to the xcircuit-dev mailing list, this fix
is to utils/fraction.c and applies only to the "magic-7.1.5" version
found only on my website.  I made promises sometime ago to fold this
stuff into the regular distribution sometime this month, but it's probably
going to be a few more weeks (plus, I've got another related bug to look
into).

						Regards,
						Tim

• Next message: DR. OZI YAKUBU: "BUSINESS PROPOSAL"
• Previous message: Conrad H Ziesler: "magscale 1 -2 bug"
• Maybe in reply to: Conrad H Ziesler: "magscale 1 -2 bug"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]