I want to create a trait that says it implements num_traits::pow::Pow - Rust.
My trait is currently defined as:
pub trait PrimeSieveTrait:
AddAssign + MulAssign + integer::Roots + FromPrimitive + ToPrimitive + PartialOrd + Copy {}
impl<T> PrimeSieveTrait for T
where T:
AddAssign + MulAssign + integer::Roots + FromPrimitive + ToPrimitive + PartialOrd + Copy {}
The types of T
I care about are u32
and u64
, which already have a .pow()
function implemented at num_traits::pow::Pow - Rust, and which return u32
and u64
. How can I modify the trait definition to get the below function working?
fn test_case<T: PrimeSieveTrait>(p: T, x: u32) -> T {
let one = T::from_u8(1).unwrap();
p.pow(x) - one
}
Rust's errors were quite helpful:
error[E0599]: no method named `pow` found for type parameter `T` in the current scope
--> src/common/divisor.rs:30:19
|
30 | let p_power = p.pow(x);
| ^^^ method not found in `T`
|
= help: items from traits can only be used if the type parameter is bounded by the trait
help: the following traits define an item `pow`, perhaps you need to restrict type parameter `T` with one of them:
|
26 | fn step_calc_divisors<T: rug::ops::Pow + PrimeSieveTrait>(divisors: &mut Vec<T>, strict_lower_bound: T, upper_bound: T, p: T, x: u32) {
| ^^^^^^^^^^^^^^^^^^
26 | fn step_calc_divisors<T: num::PrimInt + PrimeSieveTrait>(divisors: &mut Vec<T>, strict_lower_bound: T, upper_bound: T, p: T, x: u32) {
| ^^^^^^^^^^^^^^^^^
26 | fn step_calc_divisors<T: num::traits::Pow + PrimeSieveTrait>(divisors: &mut Vec<T>, strict_lower_bound: T, upper_bound: T, p: T, x: u32) {
| ^^^^^^^^^^^^^^^^^^^^^
Trying the second one worked:
pub trait PrimeSieveTrait:
AddAssign + MulAssign + DivAssign + integer::Roots + PrimInt + FromPrimitive {}
impl<T> PrimeSieveTrait for T
where T:
AddAssign + MulAssign + DivAssign + integer::Roots + PrimInt + FromPrimitive {}