Karim Belabas on Tue, 20 Jan 2004 00:02:19 +0100


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

Re: class group bug


* Igor Schein <igor@txc.com> [2004-01-19 21:57]:
> ? setrand(1582268146);
> ? bnf=bnfinit(y^12+6*y^10+31*y^8+84*y^6+159*y^4+166*y^2+1);
> ? bnr=bnrinit(bnf,4,1);
> ? rnfkummer(bnr,[1,0,0,0,0,0;0,1,0,0,0,0;0,0,1,0,0,0;0,0,0,1,0,0;0,0,0,0,2,1;0,0,0,0,0,1]);
>   ***   bug in gadd, please report

An easier test-case:
    setrand(1701015992)
    factorff(x^2+y+2, 3, y^3+y^2+2)
(same result).

A typo in FqX_split()  [ = Cantor - Zassenhaus ]: degpol(x) == 0 indeed
detects constant polynomials, _except_ 0 ... ]. Fixed in CVS.

    Karim.

P.S: Actually, 0 is in fact impossible, but the relevant check / optimization 
was forgotten.
-- 
Karim Belabas                     Tel: (+33) (0)1 69 15 57 48
Dep. de Mathematiques, Bat. 425   Fax: (+33) (0)1 69 15 60 19
Universite Paris-Sud              http://www.math.u-psud.fr/~belabas/ 
F-91405 Orsay (France)            http://pari.math.u-bordeaux.fr/  [PARI/GP]