Re: Arbitrary precision integer arithmetic

[Colin Macdonald]

On 2018-09-07 10:21 PM, Etienne Grossmann wrote:
> Hi again,
>
> sorry, I didn't express myself clearly, and I didn't know about the 
> sym() function, only the vpa() function.
>
> Let's say I want to compute 423^567 exactly.
>
> With the vpa() function, I have to specify the length of the result, as 
> in vpa (vpa(423)^567, 1500).
>
> The correct approach is to use the sym() function:  with the sym() 
> function, sym('423^567') gives me the correct answer.

Sorry I'm late to this party.  You're right; vpa is not appropriate 
here: that is generally for variable precision floating point (i.e., 
inexact with rounding).

Your example is most easily done as:

sym(423)^567

And if you need to type a large integer, use strings:

sym('12345682349871248651953253257235')^sym('123123124125')

- - - - -

intinf looks like it was great fun to write so far!  If you're into this 
sort of thing, I can definitely use help with the Symbolic package---and 
even more so with upstream SymPy that powers it.

The Interval package also involves @classes and overloading everything :)

cheers,
Colin