Frank Steffahn

Possible approach to fix #71

1

Resolves #71.

`OwningRef::map` is unsound

1

```rust fn main() { let x = OwningRef::new(Box::new(())); let z: OwningRef; { let s = "Hello World!".to_string(); let s_ref: &str = &s; let y: OwningRef = x.map(|_| &s_ref); z =...

`get_disjoint_mut` makes Miri unhappy

2

Running the doc test code, i.e. ```rs fn main() { use slotmap::*; let mut sm = SlotMap::new(); let ka = sm.insert("butter"); let kb = sm.insert("apples"); let kc = sm.insert("charlie"); sm.remove(kc);...

GATs could eliminate the need for traits for the operations

7

I haven’t studied use-cases of `typenum` too deeply, so I’m not sure how beneficial this is in the first place… and implementing this would also possibly be somewhat of a...

Still triggers UB

6

So, I wrote some random testing, and it consistently fails `cargo miri run` for me, so… have fun with it :-) ```rs #![allow(clippy::all)] use std::any::Any; use fusebox::FuseBox; use rand::random; fn...

Future API improvement ideas (thread-safety, async-support, and more)

3

With the new implementation, thread-safety shouldn't really be any problem - essentially adding the right manual `Send` and `Sync` implementations should be all that's necessary.[^1] [^1]: For `BoxScope`, the `Send`...

Soundness hole: Use-after-free through mismatching lifetimes of stream item and yielded type This is almost trivially simple to run into. Just yield something that's shorter-lived than what the place using...

It seems (me looking at macro expansions) that `yield` usage in an `async {}` block is not expanded by this macro, and that’s good - in principle - because it...

Or… “`with_mut` is still unsound” (in reference to #98). Not my title of choice though, because I’m not sure yet if I would actually blame `with_mut` here. Because the below...