Jerome Humbert


@mockersf there's some spurious CI failure, could you please re-run? Thanks!

> Those have been fixed on main, and have no chance to be cause by this PR... Also only you or Cart could rerun CI, I can just trigger bors,...

I would expect that: - a `debug_line!()` API is super convenient for checking a few small stuffs, but is limited in terms of performance and number of items it can...

That looks superb, really looking forward to try this! :) There's a few changes to examples that are not documented in the PR description. Also this is a 50+ file...

> it makes more sense for inset to create a smaller rect but the orginal f32 rect implementation has it creating a larger rect and only when a negative number...

> > Is there a reason why this shouldn't be just a third party crate? > > The previously mentioned RFC goes into reasoning: https://github.com/LegNeato/bevy-rfcs/blob/entropy/rfcs/55-deterministic_rng.md#motivation I don't see where, especially...

> The rationale being that pretty much the RNG side of the rust ecosystem has converged on rand and its ecosystem, making it the de-facto standard. This is a rationale...

Seconding @alice-i-cecile, reading the RFC I don't really see a clear advantage to a built-in solution over a 3P crate. Maybe we should move the discussion there to clarify the...

> I went ahead and added a commit that goes back to serializing entity identifiers as u64s, but via changing the implementation of `Serialize` and `Deserialize` on `Entity` instead of...