On Tue, Nov 18, 2008 at 4:30 PM, John H Palmieri <[EMAIL PROTECTED]> wrote:
>
> Does Sage know how to compute pth roots in any finite field of
> characteristic p?

There exists a finite field of char p such that Sage can compute
the p-th roots of a number:

sage: a = GF(7)(2)
sage: a.nth_root(7)
2

Same for a non-prime field:

sage: a = GF(49,'b').0
sage: a.nth_root(7)
6*b + 1

So I guess the answer to your question is probably "yes".

Incidentally, on a finite field p-th powering is an automorphism, so
there is always exactly one p-th root.  The nth_root command works
by finding the roots of the poly x^n - a.

> That is, if I have a prime number p, a finite field
> F of characteristic p, and an element b of F, will
>
> sage: b.nth_root(p)
>
> always return a value, or will it sometimes return a ValueError?

It should *always* return a value.  If it doesn't, that's a bug.

>  Does
> it depend on the field, and if so, is there a way to test whether any
> given field has an algorithm for computing pth roots?  (If I want to
> implement something which depends on such an algorithm, I want to be
> able to ignore the bad fields from the start.)

All fields should have such an algorithm.  However, one could
write a better algorithm than what is in Sage now.  E.g., if
your field F has cardinality p^n, than an algorithm to compute
the unique p-th root b of a is to just to use that for all x
in the field x^(p^n) = x, so since

   b^p = a

we have

 b = (b^p)^(p^(n-1)) = a^(p^(n-1))

I.e., compute p-th roots by raising to the power of p^(n-1).

sage: a = GF(49,'b').0
sage: a.nth_root(7)
6*b + 1
sage: a^7
6*b + 1


> I don't know what infinite fields of characteristic p Sage knows
> about, but one could ask similar questions for those: is there
> something like an 'is_perfect' method, and for which perfect fields
> does Sage know how to compute pth roots?

Infinite fields of characteristic p aren't perfect, because the definition
of perfect is that "every finite extension is separable".
In any case, Sage is I think very limited regarding non-finite characteristic
p fields, unfortunately (e.g., function fields of curves over finite fields).
This is one of those things that Magma is unusually good at.

William

--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---

Reply via email to