No i32/f64 fallback when using pow()

Question

With the following:

4.pow(2)

I get:

error: no method named `pow` found for type `_` in the current scope

I have to be more explicit about it, e.g.:

4_i32.pow(2)

With floats, I get a similar problem (and solution), where:

4.0.powf(2.0)

Must also be explicit:

4.0_f64.powf(2.0)

Answer 1

I believe this is because the fallback only happens when all the constraints on the literal have been considered (and it finds that there aren't any).

In this case, to know what the type of 4 is, it has to know what pow is. But pow is defined separately for each integral type. So to know what pow you're talking about, it has to know what the type of 4 is. But to know what the type of 4 is... and so on.