ML-KEM encapsulation is infallible, but the Kem trait always returns an EncapsError (the same probably also holds for other methods/algorithms).
If the errors were associated types, the ML-KEM implementation could specify that they're infallible and the error case wouldn't need to be handled by callers.
ML-KEM encapsulation is infallible, but the Kem trait always returns an
EncapsError(the same probably also holds for other methods/algorithms).If the errors were associated types, the ML-KEM implementation could specify that they're infallible and the error case wouldn't need to be handled by callers.