Orson Peters

> I think in order to get the same "no logic bugs" guarantee from wrapping version numbers, the program would have to make sure that it never retains dangling references....

Eureka! If we implement this we can also add a marker indicating whether a version is allowed to wraparound or not. For security sensitive application where under absolutely no circumstances...

@Robbepop I don't think that's in scope for `slotmap`. The whole point of `slotmap` is that it provides a data structure that allows for safe insertion and deletion, without deletion...

Rather than `entries`, I think a `drain_filter` method would be more inline with the rest of Rust. Would that work for you? https://doc.rust-lang.org/std/collections/hash_map/struct.HashMap.html#method.drain_filter So: ```rust fn update_players(world: &mut World, players:...

@nicoburns Well, I certainly plan on getting around to this Soon:tm:, but my Soon:tm: might not match your soon, and usually it doesn't even match my own idea of soon...

Just for the record, `SlotMap::drain_filter` would not use `drain_filter` internally, so the resulting feature would be available on stable. And I can choose whatever name I want to, I think...

I have been very busy and not always as productive the last year and a bit. I'd like to move this into slotmap 2, which I'm expecting to start working...

@starovoid The annoying part is that I've been meaning to rewrite slotmap for a while now, and that any help at this stage would be ultimately counterproductive to both parties....

Some other projects came up in the meantime, but those are either done or close to done. Soon:tm: slotmap is back on top of the stack.

Yes, this should be fixed before the next release.