AFAIK there's no optimisation based on that, just typechecking.
[0] it's interesting in that where most typecheckers restrict language semantics as they require a form of correctness proof, dialyzer is optimistic and only errors out if types are clearly inconsistent
AFAIK there's no optimisation based on that, just typechecking.
[0] it's interesting in that where most typecheckers restrict language semantics as they require a form of correctness proof, dialyzer is optimistic and only errors out if types are clearly inconsistent