Karim Belabas on Sun, 01 Feb 2004 16:13:05 +0100


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

Re: subst bug ?


* Bill Allombert [2004-02-01 11:35]:
> ? subst(a+O(b),a,c)
> %1 = c
[...]
> It looks like a bug.

Fixed in CVS. Was due to inconsistent handling of 0 t_SERs and 0 components of
polynomials. I fixed a large number of these yesterday and today, but
probably more remain [ symptom: look for gcmp0, and check whether it shouldn't
be isexactzero instead ]. From the CHANGES files

  16- problems with 0 t_SERs:
       O(y^5)/(1+x) + O(x^3)  --> O(x^3)
       O(y^5)/(1+x+O(x^3))    --> O(y^5) + O(y^5)*x + O(y^5)*x^2 + O(x^3)
       O(y) / (1+x)           --> O(y)
       O(y) * 1/(1+x)         --> O(y)
       subst(x+O(b),x,a)      --> a
   17- (0. * x) / x --> 0.E-28 * x
   18- content(0.*x) --> 0

    Karim.
-- 
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]