James Liu

A quick round of benchmarks seems to show regressions in dense iteration with a significant speedup to `World::entity` and `World::get`, not unlike the results from #3678, which also shrunk `EntityMeta`....

Done with a set of microbenchmarks relative to the current main. Most of the prior regressions have become slight improvements, or are within the noise threshold at this point. However,...

Reworked this PR to only newtype the indicies, without changing their size. This can be done in #6681 instead. Re-running microbenchmarks to ensure that this newtype does not cause the...

Microbenchmark results. Everything seems to be within the noise threshold for these benchmarks, though with some odd exceptions in busy_systems with no consistent pattern. I'd say this is mergable as...

Had to add `ArchetypeRow::new` to address it, made it pub, in line with `TableRow::new`. Should I also add a `ArchetypeRow::index` for parity?

bors retry

> I think my main concern with this would be people expecting the extracted joints to be populated and their system being scheduled for after the vec was swapped and...

Closing in favor of #6833 for the target use case.

This PR has been adopted as #6831.

@the10thWiz do you have the time to drive this PR to completion? This would be super useful in simplifying our animation code, and I'd like to see it completed. If...