Karim BELABAS on Thu, 18 Jul 2002 00:50:33 +0200 (MEST)

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

Re: more bugs

On Mon, 15 Jul 2002, Igor Schein wrote:
> On Wed, Jul 03, 2002 at 09:34:00PM -0400, Igor Schein wrote:
>> ? bnfcertify(bnfinit(x^2-40!))
>>   ***   not an integer argument in an arithmetic function
>> Works in a stable branch
> This is fixed now, but this one is broken:
> bnfcertify(bnfinit(x^2-15!))
> Infinite loop.  Again, works fine in stable branch.

Different problem. When computing a (proven) lower bound for the regulator,
there's an exhaustive enumeration of lattice vectors, which can be very
costly. The stable branch includes a cut-off point to avoid wasting too much
time in Fincke-Pohst algorithm.

I had removed that cut-off since I mistakenly thought it was there to avoid
storage problems, which I had fixed. I have put it back, together with a
small comment.

Karim Belabas                    Tel: (+33) (0)1 69 15 57 48
Dép. de Mathematiques, Bat. 425  Fax: (+33) (0)1 69 15 60 19
Université Paris-Sud             Email: Karim.Belabas@math.u-psud.fr
F-91405 Orsay (France)           http://www.math.u-psud.fr/~belabas/
PARI/GP Home Page: http://www.parigp-home.de/