1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174
//! Provides algorithms for broad phase collision detection. Specifically, implements //! Zomorodian and Edelsbrunner's hybrid algorithm using streamed segment trees, pruning and scanning, //! described in [Fast software for box intersections](https://dl.acm.org/doi/10.1145/336154.336192). //! Takes much inspiration from [the implementation in CGAL](https://github.com/CGAL/cgal/tree/master/Box_intersection_d/include/CGAL). //! //! # Examples //! ``` //! use box_intersect_ze::set::BBoxSet; //! use box_intersect_ze::boxes::Box3Df32; //! use rand_chacha::ChaCha8Rng; //! use rand::SeedableRng; //! //! // create some boxes //! let box0 = Box3Df32::new([0.0, 0.0, 0.0], [10.0, 10.0, 10.0]); //! let box1 = Box3Df32::new([5.0, 5.0, 5.0], [15.0, 15.0, 15.0]); //! let box2 = Box3Df32::new([10.0, 10.0, 10.0], [20.0, 20.0, 20.0]); //! //! // add them to a BBoxSet //! let mut boxes = BBoxSet::with_capacity(3); //! boxes.push(0, box0); //! boxes.push(1, box1); //! boxes.push(2, box2); //! boxes.sort(); // sort it in dimension 0 //! //! let mut result = Vec::with_capacity(2); // set capacity according to expected number of intersections to avoid resizing //! box_intersect_ze::intersect_ze(&boxes, &boxes, &mut result, &mut ChaCha8Rng::seed_from_u64(1234)); // get the intersections //! //! assert!(result.contains(&(1,0))); //! assert!(result.contains(&(2,1))); //! assert!(!result.contains(&(2,0))); //! assert!(!result.contains(&(0,2))); //! ``` use boxes::BBox; use set::BBoxSet; use crate::internals::{hybrid, one_way_scan, two_way_scan}; pub mod boxes; pub mod internals; mod median; pub mod set; /// Trait for box boundary types pub trait HasInfinity { /// Value representing negative infinity const NINFTY: Self; /// Value representing positive infinity const INFTY: Self; } /// Trait for random number generator used in [`intersect_ze`] for approximate median calculation pub trait Rng { /// Returns a random `usize` between 0 (inclusive) and `high` (exclusive) fn rand_usize(&mut self, high: usize) -> usize; } #[cfg(feature = "rand-crate")] impl<R> Rng for R where R: rand::Rng, { fn rand_usize(&mut self, max: usize) -> usize { self.gen_range(0..max) } } /// Finds all intersections between boxes in `a` and `b` using Zomorodian and Edelsbrunner's /// hybrid algorithm (streamed segment trees pruned with a cutoff). /// * `a` and `b` may be either the same or distinct [`BBoxSet`]s and must be sorted before calling. /// * `out` will contain pairs of `ID`s of intersecting boxes. /// Choose capacity according to the number of intersections you expect to avoid resizing. /// * `rand` must be a random number generator implementing the [`Rng`] trait. (used for approximate median selection) pub fn intersect_ze<B, ID, R>( a: &BBoxSet<B, ID>, b: &BBoxSet<B, ID>, out: &mut Vec<(ID, ID)>, rand: &mut R, ) where B: BBox, B::Num: PartialOrd + HasInfinity, ID: PartialOrd + Copy, R: Rng, { const CUTOFF: usize = 1000; // should give reasonable performance for up to 100,000 boxes intersect_ze_custom::<B, ID, R, CUTOFF>(a, b, out, rand); } /// Like `intersect_ze` but with a customizable cutoff. pub fn intersect_ze_custom<B, ID, R, const CUTOFF: usize>( a: &BBoxSet<B, ID>, b: &BBoxSet<B, ID>, out: &mut Vec<(ID, ID)>, rand: &mut R, ) where B: BBox, ID: PartialOrd + Copy, B::Num: PartialOrd + HasInfinity, ID: PartialEq, R: Rng, { let same = a as *const _ == b as *const _; if same { // one tree is enough to have every box represented as both an interval and a point hybrid::<B, ID, R, CUTOFF>(a, a, B::Num::NINFTY, B::Num::INFTY, B::DIM - 1, out, rand); } else { // need two trees so that every box is represented as both an interval and a point hybrid::<B, ID, R, CUTOFF>(a, b, B::Num::NINFTY, B::Num::INFTY, B::DIM - 1, out, rand); hybrid::<B, ID, R, CUTOFF>(b, a, B::Num::NINFTY, B::Num::INFTY, B::DIM - 1, out, rand); } } /// Finds all intersections between boxes in `a` and `b` using a scanning algorithm. /// Should perform reasonably up to approximately 1,000 boxes /// * `a` and `b` may be either the same or distinct [`BBoxSet`]s and must be sorted before calling. /// * `out` will contain pairs of `ID`s of intersecting boxes. pub fn intersect_scan<B, ID>(a: &BBoxSet<B, ID>, b: &BBoxSet<B, ID>, out: &mut Vec<(ID, ID)>) where B: BBox, ID: Copy + PartialOrd, { let same = a as *const _ == b as *const _; // check if a and b refer to the same BBoxSet if same { one_way_scan(a, b, B::DIM - 1, out); } else { two_way_scan(a, b, out); } } /// Finds box intersections by checking every box in `a` against every box in `b`. /// Performs well for on the order of 100 boxes. *O*(*n^2*) /// * `a` and `b` may be either the same or distinct [`BBoxSet`]s /// * `out` will contain pairs of `ID`s of intersecting boxes. pub fn intersect_brute_force<B, ID>(a: &BBoxSet<B, ID>, b: &BBoxSet<B, ID>, out: &mut Vec<(ID, ID)>) where B: BBox, ID: Copy, { let same = a as *const _ == b as *const _; // check if a and b refer to the same BBoxSet if same { // avoid duplicate intersections let mut start = 1; for &(bbox, id) in &a.boxes { for idx in start..a.boxes.len() { let (bbox2, id2) = a.boxes[idx]; if bbox.intersects(&bbox2) { out.push((id, id2)); } } start += 1; } } else { for &(bbox, id) in &a.boxes { for &(bbox2, id2) in &b.boxes { if bbox.intersects(&bbox2) { out.push((id, id2)); } } } } } impl HasInfinity for f32 { const NINFTY: Self = f32::NEG_INFINITY; const INFTY: Self = f32::INFINITY; } impl HasInfinity for f64 { const NINFTY: Self = f64::NEG_INFINITY; const INFTY: Self = f64::INFINITY; } #[cfg(test)] mod tests;