Bill Allombert on Mon, 17 Jan 2005 11:11:15 +0100


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: Gaussian Integer factorisation at right angles


On Sat, Jan 15, 2005 at 01:29:45PM +0100, Karim Belabas wrote:
> Anyway, I have found the problem. For some reason, my implementation
> expected qfbsolve to always return a normalized solution (in north-east
> quadrant).

I suppose we could/should canonicalize the output of qfbsolve but
this particular case is a non-generic : there are 8 solutions instead of
4.

One day, I should add support for real quadratic forms in qfbsolve but
there are no polynomial-time algorithms so it is less fun.

Cheers,
Bill.