Emilia Bopp

I am starting to think that dynamic switching could also be removed in favour of using simpler idioms in user code. Thoughts?

Turns out lifetime issues are indeed hairy. It seems also like this would require making the log explicit, as it can't be a thread-local variable, when its lifetime is limited...

It is well possible that the majority of use cases for this simply cannot deal with the ownership semantics at compile-time. So consider this an experiment for now. I tried...

There is a pull request for the refactor steps now, see #4. And perhaps you want to check out my work in progress branch concerning the non-static lifetimes using an...

I got distracted and started doing other things ;)

Also maybe a Barnes-Hut tree force calculation is not the most minimal way to show off the functionality of this crate. Maybe a nearest neighbour search would be better.

Looks like the performance penalty is negligible. No need to introduce an unsafe API for one percent performance gain imho. The check you are proposing is more about improving error...

Why would it have to be anything more than a barebones enum? And what precisely do you mean by re-export?

#7 must be addressed first.

Before #42 is resolved, this can not be solved nicely. Nonetheless, it is possible to work around the problem using the above implementation strategy.