Georg Semmler

> If we had some sort of internal trait that exposed a const fn identifier_fn() -> usize it should be possible to insert something like: I fear that won't work...

I must admit that I still do not see how you can go from this: > ```rust > const _CONSTRAINT: () = assert!( > !are_equal(T1::SQL_NAME, T2::SQL_NAME), > "cannot use identifier...

> I'm still honestly trying to wrap my head around why we need Peano numbers here, and how the traits all interlock. I was trying to remove the table equals...

That's all fine and good but again I still cannot see how you would implement that const panic thing. At some point you must use generics from some outer trait...

@lsunsi It's not that we are not aware of this issue, it's just that we are not aware of a general solution to this problem. We would be happy to...

The current implementation just generates a call for all tables for several reasons: * It's much simpler to implement * It also covers the case where you manually join tables...

From my point of view it would be ok to support a "none" option there. That written: I also need to correct a statement made above. It seems like it...

@PratikFandade Sure, just let us know if you need some help there. Also expect to handle some rather complex trait relations there.

@prkbuilds Sorry for taking that long to answer. I've not forgot about this, but I had not the capacity yet to investigate why this happens. It might help to have...

Thanks for reporting. That's definitively a bug. I think we should be able to fix that by just checking here (https://github.com/diesel-rs/diesel/blob/381be195688db339fe2927e49bc818ab86754dd9/diesel/src/pg/types/ranges.rs#L83) if `flags` contains `RangeFlags::Empty` and if that's the case...