The code example will work even if `u` is only known at runtime. That's because the inner switch is not matching on `u`, it's matching on `ab`, which is known at compile time due to the use of `inline`.
That may be confusing, but basically `inline` is generating different code for the branches .a and .b, so in those cases the value of `ab` is known at compile time. So, the inner switch is running at compile time too. In the .a branch it just turns into a call to handle_a(), and in the .b branch it turns into a call to handle_b().
Agreed. Zig's approach re-uses the existing machinery of the language far more than C++ templates do. Another example of this is that Zig has almost no restrictions on what kinds of values can be `comptime` parameters. In C++, "non-type template parameters" are restricted to a small subset of types (integers, enums, and a few others). Rust's "const generics" are even more restrictive: only integers for now.
In Zig I can pass an entire struct instance full of config values as a single comptime parameter and thread it anywhere in my program. The big difference here is that when you treat compile-time programming as a "special" thing that is supported completely differently in the language, you need to add these features in a painfully piecemeal way. Whereas if it's just re-using the machinery already in place in your language, these restrictions don't exist and your users don't need to look up what values can be comptime values...they're just another kind of thing I pass to functions, so "of course" I can pass a struct instance.
That may be confusing, but basically `inline` is generating different code for the branches .a and .b, so in those cases the value of `ab` is known at compile time. So, the inner switch is running at compile time too. In the .a branch it just turns into a call to handle_a(), and in the .b branch it turns into a call to handle_b().