@jasonish how do you recommend doing this ?
We will need the same structure and code for u8, u16, u32 and u64

I guess one could use an enum?

enum UintValue {
    U8(u8),
    U16(u6),
    U32(u32),
}

struct DetectUintData {
    pub val: UintValue,
    pub valrange: UintValue,
}

it will at least reduce non-common code that is only size specific.

With this, val could be u8 and valrange u16 ...
I will look into it a bit more, see how other crates handled it...

I see that num-integer (used by suricata) uses macros for such as case cf https://github.com/rust-num/num-integer/blob/master/src/roots.rs#L165

@jasonish is not there a derive thing like suricata_derive::AppLayerEvent ?

Or maybe use crate https://docs.rs/num/0.4.0/num/integer/trait.Integer.html ?

I see that num-integer (used by suricata) uses macros for such as case cf https://github.com/rust-num/num-integer/blob/master/src/roots.rs#L165

@jasonish is not there a derive thing like suricata_derive::AppLayerEvent ?

No, we'd have to write that.

Or maybe use crate https://docs.rs/num/0.4.0/num/integer/trait.Integer.html ?

If these fit your needs, then I think they are good to use. We already have them in our deps. I don't think I've used them myself though.

Managing something with rust generics... let's talk in next version of PR ;-)