On Fri, 13 Oct 2006 23:32:09 -0700, didier deshommes <[EMAIL PROTECTED]>
wrote:
> Currently abs() is not defined for integers with a modulus:
I don't think it should be defined. Mathematically, it doesn't makes
sense to
write |n| for n an integer modulo m.
To add strength to my argument that this shouldn't be defined in SAGE, note
that in both GP/PARI and Magma abs is not defined on integers modulo n:
abs(Mod(-3,10)) in PARI:
GP/PARI ERROR:
*** abs: incorrect type in gabs.
Abs(Integers(10)!(-3));
^
Runtime error in 'Abs': Bad argument types
Argument types given: RngIntResElt
You can get what you are thinking of as "the absolute value" using the
lift method:
sage: Mod(7,10).lift()
7
Sorry to bounce your patch -- I hate not incorporating things people send
me,
but in this case I think it is not good. It might even make sense to
have a patch that defines __abs__ for integers modulo N to raise an
ArithmeticError that explains that abs shouldn't make sense in that
context,
and the user should call lift instead.
-- William
> sage: abs(mod(7,10))
> ---------------------------------------------------------------------------
> <type 'exceptions.TypeError'> Traceback (most recent call
> last)
> <type 'exceptions.TypeError'>: bad operand type for abs():
> 'integer_mod.IntegerMod_int'
>
> Here is a diff against integer_mod.pyx . The change is small enough so
> that I didn't need to make a full bundle (I'm just getting the hang of
> hg):
>
> diff -r 5eba42221f8a -r 6a378a7db7ec sage/rings/integer_mod.pyx
> --- a/sage/rings/integer_mod.pyx Thu Oct 05 23:44:44 2006 -0700
> +++ b/sage/rings/integer_mod.pyx Fri Oct 13 23:17:46 2006 -0700
> @@ -944,6 +944,15 @@ cdef class IntegerMod_int(IntegerMod_abs
> def __hash__(self):
> return hash(self.ivalue)
>
> + def __abs__(self):
> + """
> + EXAMPLES:
> + sage: abs(mod(3,10))
> + 3
> + sage: abs(mod(-3,10))
> + 7
> + """
> + return self.ivalue
> ### End of class
>
>
> Cheers,
> didier
>
> >
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~----------~----~----~----~------~----~------~--~---
