Violet Leonard

> Just because we can do something doesn't mean we should. Totally fair, probably it's a good idea to table this one for now until someone actually asks for a...

I've thought about that, and actually do have a use case where I could possibly use such a thing, however there are additional complexities. If we're just talking about a...

Right, and that requires the same type of runtime uniqueness checking. It does seem like multiple unique borrows from a hashmap is a bit of an API hole that qcell...

I notice there's a someday-maybe label, that would probably be appropriate here.

After looking into this a little, here's some commentary: LCell will work as-is. TLCell is basically completely off the table, at least with the current state of Rust, as thread...

One thing maybe worth mentioning is it's relatively easy to make a global collection without std - if it's fixed size. So we could say, for example, "on no_std, TCell...

I've been experimenting with something which is a possible solution to no_std TCell. [https://github.com/geeklint/typeid-set](https://github.com/geeklint/typeid-set) is a data structure which could be used in place of the current Mutex + CondVar,...

That totally makes sense to me, I think I will work on deploying this as its own crate, and may submit a PR here with the optional dependency at some...

Getting around to working on this, I've published the data structure as [exclusion-set](https://docs.rs/exclusion-set/latest/exclusion_set/), and I'm working on a PR for the optional dep. One thing I've noticed is it might...

Correct, it's my understanding that it should only effect people testing qcell itself, not users of the crate. The greatest value is probably just the examples in the doc comment...