Return distances from KDTree::within function

[adamthedash]

Hi, thanks for creating this crate!
For my use case (swarm simulation), I'm searching for nearby entities using KDTree::within. However as it only returns the indices, I'm re-calculating the distances of neighbours to use downstream. I see internally that KDTree::within is already calculating it for candidate points, so was wondering if they could also be returned alongside the indices?

[kylebarron]

Hello 👋

That sounds like you probably ideally want to use nearest neighbors instead of within. This is tracked by #97.

The RTree implementation already has the nearest neighbor implementation, but an improvement will be to expose it as a rust Iterator instead of collecting into a Vec, see #98 for a WIP but not yet functional implementation.

In this case, the item yielded by the iterator includes both the id and the (squared) distance:

geo-index/src/rtree/trait.rs

Lines 248 to 249 in 9ca258b

	id: usize,
	dist: N,

I think this approach would work the best for you as well: to implement an iterator-based nearest-neighbor search and then you can stop iteration once you've hit a certain number or distance.