fastquadtree 0.4.1__tar.gz → 0.5.0__tar.gz

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/.github/workflows/release.yml

@@ -2,8 +2,7 @@ name: Publish to PyPI
 
 on:
   push:
-    tags:
-      - "v*"
+    tags: ["v*"]
   release:
     types: [published]
   workflow_dispatch:
@@ -18,21 +17,41 @@ jobs:
         uses: actions/setup-python@v5
         with:
           python-version: "3.10"
-      - name: Build with maturin
+
+      - name: Create venv for maturin develop
+        run: python -m venv .venv
+
+      - name: Build with maturin & install into .venv
         uses: PyO3/maturin-action@v1
         with:
-          command: build
+          command: develop
+          args: --release
           manylinux: manylinux2014
-      - name: Install Dependencies
-        run: pip install pytest pytest-cov
-      - name: Test with pytest
-        run: pytest --cov --cov-branch --cov-report=xml
+
+      - name: Install Python test deps into .venv
+        run: |
+          . .venv/bin/activate
+          pip install -U pip pytest pytest-cov
+
+      - name: Run Python tests
+        run: |
+          . .venv/bin/activate
+          pytest
+
+      - name: Install Rust toolchain
+        uses: dtolnay/rust-toolchain@stable
+
+      - name: Run Rust tests
+        run: cargo test
+
       - name: Upload coverage to Codecov
         uses: codecov/codecov-action@v5
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
+
   build:
     name: Build wheels
+    needs: test
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
@@ -45,8 +64,7 @@ jobs:
             args: "--skip-existing"
     steps:
       - uses: actions/checkout@v4
-      - name: Set up Python
-        uses: actions/setup-python@v5
+      - uses: actions/setup-python@v5
         with:
           python-version: "3.10"
       - name: Build and publish with maturin
@@ -56,4 +74,4 @@ jobs:
         with:
           command: publish
           args: ${{ matrix.args }}
-          manylinux: manylinux2014
+          manylinux: manylinux2014

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/.pre-commit-config.yaml

@@ -31,16 +31,7 @@ repos:
       # This both runs pytest and generates .coverage data
       - id: pytest
         name: pytest (under coverage)
-        entry: bash -c "coverage run -m pytest -q"
+        entry: pytest
         language: system
         pass_filenames: false
-        always_run: true
-
-      # 4) Coverage threshold check
-      # Change FAIL_UNDER to your target percent
-      - id: coverage-report
-        name: coverage report
-        entry: bash -c 'coverage report --omit="tests/*" --fail-under=85'
-        language: system
-        pass_filenames: false
-        always_run: true
+        always_run: true

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/Cargo.lock

@@ -22,7 +22,7 @@ checksum = "2fd1289c04a9ea8cb22300a459a72a385d7c73d3259e2ed7dcb2af674838cfa9"
 
 [[package]]
 name = "fastquadtree"
-version = "0.4.1"
+version = "0.5.0"
 dependencies = [
  "pyo3",
 ]

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "fastquadtree"
-version = "0.4.1"
+version = "0.5.0"
 edition = "2021"
 readme = "README.md"
 
@@ -9,3 +9,9 @@ crate-type = ["rlib", "cdylib"]
 
 [dependencies]
 pyo3 = { version = "0.21", features = ["extension-module", "abi3-py38"] }
+
+[profile.release]
+opt-level = 3
+lto = "thin"
+codegen-units = 1
+panic = "abort"

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: fastquadtree
-Version: 0.4.1
+Version: 0.5.0
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3 :: Only
 Classifier: Programming Language :: Rust
@@ -37,7 +37,6 @@ Project-URL: Issues, https://github.com/Elan456/fastquadtree/issues
 [![Rust core via PyO3](https://img.shields.io/badge/Rust-core%20via%20PyO3-orange)](https://pyo3.rs/)
 [![Built with maturin](https://img.shields.io/badge/Built%20with-maturin-1f6feb)](https://www.maturin.rs/)
 [![Code style: Black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)
-[![Type checking: mypy](https://img.shields.io/badge/type%20checking-mypy-2a6db2)](http://mypy-lang.org/)
 
 
 ![Interactive_V2_Screenshot](https://raw.githubusercontent.com/Elan456/fastquadtree/main/assets/interactive_v2_screenshot.png)
@@ -49,6 +48,29 @@ Rust-optimized quadtree with a simple Python API.
 - Python ≥ 3.8
 - Import path: `from fastquadtree import QuadTree`
 
+## Benchmarks
+
+fastquadtree **outperforms** all other quadtree Python packages, including the Rtree spatial index.
+
+### Library comparison
+
+![Total time](https://raw.githubusercontent.com/Elan456/fastquadtree/main/assets/quadtree_bench_time.png)
+![Throughput](https://raw.githubusercontent.com/Elan456/fastquadtree/main/assets/quadtree_bench_throughput.png)
+
+### Summary (largest dataset, PyQtree baseline)
+- Points: **500,000**, Queries: **500**
+--------------------
+- Fastest total: **fastquadtree** at **1.591 s**
+
+| Library | Build (s) | Query (s) | Total (s) | Speed vs PyQtree |
+|---|---:|---:|---:|---:|
+| fastquadtree | 0.165 | 1.427 | 1.591 | 5.09× |
+| Rtree        | 1.320 | 2.369 | 3.688 | 2.20× |
+| PyQtree      | 2.687 | 5.415 | 8.102 | 1.00× |
+| nontree-QuadTree | 1.284 | 9.891 | 11.175 | 0.73× |
+| quads        | 2.346 | 10.129 | 12.475 | 0.65× |
+| e-pyquadtree | 1.795 | 11.855 | 13.650 | 0.59× |
+
 ## Install
 
 ```bash
@@ -109,25 +131,7 @@ print(f"Deleted player: {deleted}")  # True
 
 You can keep the tree pure and manage your own id → object map, or let the wrapper manage it.
 
-**Option A: Manage your own map**
-
-```python
-from fastquadtree import QuadTree
-
-qt = QuadTree((0, 0, 1000, 1000), capacity=16)
-objects: dict[int, object] = {}
-
-def add(obj) -> int:
-    obj_id = qt.insert(obj.position)  # auto id
-    objects[obj_id] = obj
-    return obj_id
-
-# Later, resolve ids back to objects
-ids = [obj_id for (obj_id, x, y) in qt.query((100, 100, 300, 300))]
-selected = [objects[i] for i in ids]
-```
-
-**Option B: Ask the wrapper to track objects**
+**Wrapper Managed Objects**
 
 ```python
 from fastquadtree import QuadTree
@@ -137,7 +141,7 @@ qt = QuadTree((0, 0, 1000, 1000), capacity=16, track_objects=True)
 # Store the object alongside the point
 qt.insert((25, 40), obj={"name": "apple"})
 
-# Ask for Item objects so you can access .obj lazily
+# Ask for Item objects within a bounding box
 items = qt.query((0, 0, 100, 100), as_items=True)
 for it in items:
     print(it.id, it.x, it.y, it.obj)
@@ -151,7 +155,7 @@ qt.attach(123, my_object)  # binds object to id 123
 
 ## API
 
-### `QuadTree(bounds, capacity, *, max_depth=None, track_objects=False, start_id=1)`
+### `QuadTree(bounds, capacity, max_depth=None, track_objects=False, start_id=1)`
 
 * `bounds` — tuple `(min_x, min_y, max_x, max_y)` defines the 2D area covered by the quadtree
 * `capacity` — max number of points kept in a leaf before splitting
@@ -204,30 +208,8 @@ Full docs are in the docstrings of the [Python Shim](pysrc/fastquadtree/__init__
 * For fastest local runs, use `maturin develop --release`.
 * The wrapper keeps Python overhead low: raw tuple results by default, `Item` wrappers only when requested.
 
-## Benchmarks
-
-fastquadtree outperforms all other quadtree python packages (at least all the ones I could find and install via pip.)
-
-### Library comparison
-
-![Total time](https://raw.githubusercontent.com/Elan456/fastquadtree/main/assets/quadtree_bench_time.png)
-![Throughput](https://raw.githubusercontent.com/Elan456/fastquadtree/main/assets/quadtree_bench_throughput.png)
-
-### Summary (largest dataset, PyQtree baseline)
-- Points: **500,000**, Queries: **500**
-- Fastest total: **fastquadtree** at **2.207 s**
-
-| Library | Build (s) | Query (s) | Total (s) | Speed vs PyQtree |
-|---|---:|---:|---:|---:|
-| fastquadtree  | 0.321 | 1.885 | 2.207 | 4.27× |
-| Rtree        | 1.718 | 4.376 | 6.095 | 1.55× |
-| nontree-QuadTree | 1.617 | 7.643 | 9.260 | 1.02× |
-| PyQtree      | 4.349 | 5.082 | 9.431 | 1.00× |
-| quads        | 3.874 | 9.058 | 12.932 | 0.73× |
-| e-pyquadtree | 2.732 | 10.598 | 13.330 | 0.71× |
-| Brute force  | 0.019 | 19.986 | 20.005 | 0.47× |
 
-### Native vs Shim
+### Native vs Shim Benchmark
 
 **Setup**
 - Points: 500,000
@@ -277,7 +259,7 @@ Check the CLI arguments for the cross-library benchmark in `benchmarks/quadtree_
 Allowed. For k-nearest, duplicates are de-duplicated by id. For range queries you will see every inserted point.
 
 **Can I delete items from the quadtree?**
-Yes! Use `delete(id, xy)` to remove specific items. You must provide both the ID and exact location for precise deletion. This handles cases where multiple items exist at the same location. If you're using `track_objects=True`, you can also use `delete_by_object(obj, xy)` for convenient object-based deletion with O(1) lookup. The tree automatically merges nodes when item counts drop below capacity.
+Yes! Use `delete(id, xy)` to remove specific items. You must provide both the ID and exact location for precise deletion. This handles cases where multiple items exist at the same location. If you're using `track_objects=True`, you can also use `delete_by_object(obj)` for convenient object-based deletion with O(1) lookup. The tree automatically merges nodes when item counts drop below capacity.
 
 **Can I store rectangles or circles?**
 The core stores points. To index objects with extent, insert whatever representative point you choose. For rectangles you can insert centers or build an AABB tree separately.

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/README.md

@@ -14,7 +14,6 @@
 [![Rust core via PyO3](https://img.shields.io/badge/Rust-core%20via%20PyO3-orange)](https://pyo3.rs/)
 [![Built with maturin](https://img.shields.io/badge/Built%20with-maturin-1f6feb)](https://www.maturin.rs/)
 [![Code style: Black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)
-[![Type checking: mypy](https://img.shields.io/badge/type%20checking-mypy-2a6db2)](http://mypy-lang.org/)
 
 
 ![Interactive_V2_Screenshot](https://raw.githubusercontent.com/Elan456/fastquadtree/main/assets/interactive_v2_screenshot.png)
@@ -26,6 +25,29 @@ Rust-optimized quadtree with a simple Python API.
 - Python ≥ 3.8
 - Import path: `from fastquadtree import QuadTree`
 
+## Benchmarks
+
+fastquadtree **outperforms** all other quadtree Python packages, including the Rtree spatial index.
+
+### Library comparison
+
+![Total time](https://raw.githubusercontent.com/Elan456/fastquadtree/main/assets/quadtree_bench_time.png)
+![Throughput](https://raw.githubusercontent.com/Elan456/fastquadtree/main/assets/quadtree_bench_throughput.png)
+
+### Summary (largest dataset, PyQtree baseline)
+- Points: **500,000**, Queries: **500**
+--------------------
+- Fastest total: **fastquadtree** at **1.591 s**
+
+| Library | Build (s) | Query (s) | Total (s) | Speed vs PyQtree |
+|---|---:|---:|---:|---:|
+| fastquadtree | 0.165 | 1.427 | 1.591 | 5.09× |
+| Rtree        | 1.320 | 2.369 | 3.688 | 2.20× |
+| PyQtree      | 2.687 | 5.415 | 8.102 | 1.00× |
+| nontree-QuadTree | 1.284 | 9.891 | 11.175 | 0.73× |
+| quads        | 2.346 | 10.129 | 12.475 | 0.65× |
+| e-pyquadtree | 1.795 | 11.855 | 13.650 | 0.59× |
+
 ## Install
 
 ```bash
@@ -86,25 +108,7 @@ print(f"Deleted player: {deleted}")  # True
 
 You can keep the tree pure and manage your own id → object map, or let the wrapper manage it.
 
-**Option A: Manage your own map**
-
-```python
-from fastquadtree import QuadTree
-
-qt = QuadTree((0, 0, 1000, 1000), capacity=16)
-objects: dict[int, object] = {}
-
-def add(obj) -> int:
-    obj_id = qt.insert(obj.position)  # auto id
-    objects[obj_id] = obj
-    return obj_id
-
-# Later, resolve ids back to objects
-ids = [obj_id for (obj_id, x, y) in qt.query((100, 100, 300, 300))]
-selected = [objects[i] for i in ids]
-```
-
-**Option B: Ask the wrapper to track objects**
+**Wrapper Managed Objects**
 
 ```python
 from fastquadtree import QuadTree
@@ -114,7 +118,7 @@ qt = QuadTree((0, 0, 1000, 1000), capacity=16, track_objects=True)
 # Store the object alongside the point
 qt.insert((25, 40), obj={"name": "apple"})
 
-# Ask for Item objects so you can access .obj lazily
+# Ask for Item objects within a bounding box
 items = qt.query((0, 0, 100, 100), as_items=True)
 for it in items:
     print(it.id, it.x, it.y, it.obj)
@@ -128,7 +132,7 @@ qt.attach(123, my_object)  # binds object to id 123
 
 ## API
 
-### `QuadTree(bounds, capacity, *, max_depth=None, track_objects=False, start_id=1)`
+### `QuadTree(bounds, capacity, max_depth=None, track_objects=False, start_id=1)`
 
 * `bounds` — tuple `(min_x, min_y, max_x, max_y)` defines the 2D area covered by the quadtree
 * `capacity` — max number of points kept in a leaf before splitting
@@ -181,30 +185,8 @@ Full docs are in the docstrings of the [Python Shim](pysrc/fastquadtree/__init__
 * For fastest local runs, use `maturin develop --release`.
 * The wrapper keeps Python overhead low: raw tuple results by default, `Item` wrappers only when requested.
 
-## Benchmarks
-
-fastquadtree outperforms all other quadtree python packages (at least all the ones I could find and install via pip.)
-
-### Library comparison
-
-![Total time](https://raw.githubusercontent.com/Elan456/fastquadtree/main/assets/quadtree_bench_time.png)
-![Throughput](https://raw.githubusercontent.com/Elan456/fastquadtree/main/assets/quadtree_bench_throughput.png)
-
-### Summary (largest dataset, PyQtree baseline)
-- Points: **500,000**, Queries: **500**
-- Fastest total: **fastquadtree** at **2.207 s**
-
-| Library | Build (s) | Query (s) | Total (s) | Speed vs PyQtree |
-|---|---:|---:|---:|---:|
-| fastquadtree  | 0.321 | 1.885 | 2.207 | 4.27× |
-| Rtree        | 1.718 | 4.376 | 6.095 | 1.55× |
-| nontree-QuadTree | 1.617 | 7.643 | 9.260 | 1.02× |
-| PyQtree      | 4.349 | 5.082 | 9.431 | 1.00× |
-| quads        | 3.874 | 9.058 | 12.932 | 0.73× |
-| e-pyquadtree | 2.732 | 10.598 | 13.330 | 0.71× |
-| Brute force  | 0.019 | 19.986 | 20.005 | 0.47× |
 
-### Native vs Shim
+### Native vs Shim Benchmark
 
 **Setup**
 - Points: 500,000
@@ -254,7 +236,7 @@ Check the CLI arguments for the cross-library benchmark in `benchmarks/quadtree_
 Allowed. For k-nearest, duplicates are de-duplicated by id. For range queries you will see every inserted point.
 
 **Can I delete items from the quadtree?**
-Yes! Use `delete(id, xy)` to remove specific items. You must provide both the ID and exact location for precise deletion. This handles cases where multiple items exist at the same location. If you're using `track_objects=True`, you can also use `delete_by_object(obj, xy)` for convenient object-based deletion with O(1) lookup. The tree automatically merges nodes when item counts drop below capacity.
+Yes! Use `delete(id, xy)` to remove specific items. You must provide both the ID and exact location for precise deletion. This handles cases where multiple items exist at the same location. If you're using `track_objects=True`, you can also use `delete_by_object(obj)` for convenient object-based deletion with O(1) lookup. The tree automatically merges nodes when item counts drop below capacity.
 
 **Can I store rectangles or circles?**
 The core stores points. To index objects with extent, insert whatever representative point you choose. For rectangles you can insert centers or build an AABB tree separately.

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/benchmarks/quadtree_bench/engines.py

@@ -247,7 +247,7 @@ def get_engines(
         "fastquadtree": _create_fastquadtree_engine(bounds, max_points, max_depth),
         "e-pyquadtree": _create_e_pyquadtree_engine(bounds, max_points, max_depth),
         "PyQtree": _create_pyqtree_engine(bounds, max_points, max_depth),
-     #    "Brute force": _create_brute_force_engine(bounds, max_points, max_depth),  # Brute force doesn't scale well on the graphs so omit it from the main set
+        #    "Brute force": _create_brute_force_engine(bounds, max_points, max_depth),  # Brute force doesn't scale well on the graphs so omit it from the main set
     }
 
     # Optional engines (only include if import succeeded)

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/benchmarks/quadtree_bench/runner.py

@@ -34,7 +34,7 @@ class BenchmarkConfig:
         """Generate experiment point sizes."""
         self.experiments = [2, 4, 8, 16]
         while self.experiments[-1] < self.max_experiment_points:
-            self.experiments.append(int(self.experiments[-1] * 1.5))
+            self.experiments.append(int(self.experiments[-1] * 2))
         if self.experiments[-1] > self.max_experiment_points:
             self.experiments[-1] = self.max_experiment_points
 

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/pyproject.toml

@@ -39,3 +39,18 @@ module-name = "fastquadtree._native"
 # Choose a wide wheel tag for Linux. Adjust if your build images are newer.
 # Options include: manylinux2014, manylinux_2_28, musllinux_1_2, or "off".
 compatibility = "manylinux2014"
+
+[tool.pytest.ini_options]
+addopts = "--cov=fastquadtree --cov-branch --cov-report=xml --cov-fail-under=95"
+testpaths = ["tests"]  # still run tests
+
+[tool.coverage.run]
+source = ["fastquadtree"]
+omit = [
+  "tests/*",
+]
+
+[tool.coverage.report]
+omit = [
+  "tests/*",
+]

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/pysrc/fastquadtree/_bimap.py

@@ -80,11 +80,11 @@ class BiMap:
         """
         removed = None
         # Remove by id first
-        if self._id_to_item.get(item.id) is item:
-            removed = self._id_to_item.pop(item.id)
+        removed = self._id_to_item.pop(item.id)
+
         # Remove by obj side
         obj = item.obj
-        if obj is not None and self._objid_to_item.get(id(obj)) is item:
+        if obj is not None:
             self._objid_to_item.pop(id(obj), None)
             removed = removed or item
         return removed

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/src/geom.rs

@@ -1,15 +1,15 @@
 #[derive(Copy, Clone, Debug, PartialEq, Default)]
 pub struct Point {
-    pub x: f64,
-    pub y: f64,
+    pub x: f32,
+    pub y: f32,
 }
 
 #[derive(Copy, Clone, Debug, PartialEq, Default)]
 pub struct Rect {
-    pub min_x: f64,
-    pub min_y: f64,
-    pub max_x: f64,
-    pub max_y: f64,
+    pub min_x: f32,
+    pub min_y: f32,
+    pub max_x: f32,
+    pub max_y: f32,
 }
 
 impl Rect {
@@ -23,7 +23,7 @@ impl Rect {
     }
 }
 
-pub fn dist_sq_point_to_rect(p: &Point, r: &Rect) -> f64 {
+pub fn dist_sq_point_to_rect(p: &Point, r: &Rect) -> f32 {
     let dx = if p.x < r.min_x {
         r.min_x - p.x
     } else if p.x > r.max_x {
@@ -43,7 +43,7 @@ pub fn dist_sq_point_to_rect(p: &Point, r: &Rect) -> f64 {
     dx * dx + dy * dy
 }
 
-pub fn dist_sq_points(a: &Point, b: &Point) -> f64 {
+pub fn dist_sq_points(a: &Point, b: &Point) -> f32 {
     let dx = a.x - b.x;
     let dy = a.y - b.y;
     dx * dx + dy * dy

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/src/lib.rs

@@ -8,7 +8,7 @@ pub use crate::quadtree::{Item, QuadTree};
 
 use pyo3::prelude::*;
 
-fn item_to_tuple(it: Item) -> (u64, f64, f64) {
+fn item_to_tuple(it: Item) -> (u64, f32, f32) {
     (it.id, it.point.x, it.point.y)
 }
 
@@ -20,7 +20,7 @@ pub struct PyQuadTree {
 #[pymethods]
 impl PyQuadTree {
     #[new]
-    pub fn new(bounds: (f64, f64, f64, f64), capacity: usize, max_depth: Option<usize>) -> Self {
+    pub fn new(bounds: (f32, f32, f32, f32), capacity: usize, max_depth: Option<usize>) -> Self {
         let (min_x, min_y, max_x, max_y) = bounds;
         let rect = Rect { min_x, min_y, max_x, max_y };
         let inner = match max_depth {
@@ -30,17 +30,17 @@ impl PyQuadTree {
         Self { inner }
     }
 
-    pub fn insert(&mut self, id: u64, xy: (f64, f64)) -> bool {
+    pub fn insert(&mut self, id: u64, xy: (f32, f32)) -> bool {
         let (x, y) = xy;
         self.inner.insert(Item { id, point: Point { x, y } })
     }
 
-    pub fn delete(&mut self, id: u64, xy: (f64, f64)) -> bool {
+    pub fn delete(&mut self, id: u64, xy: (f32, f32)) -> bool {
         let (x, y) = xy;
         self.inner.delete(id, Point { x, y })
     }
 
-    pub fn query(&self, rect: (f64, f64, f64, f64)) -> Vec<(u64, f64, f64)> {
+    pub fn query(&self, rect: (f32, f32, f32, f32)) -> Vec<(u64, f32, f32)> {
         let (min_x, min_y, max_x, max_y) = rect;
         self.inner
             .query(Rect { min_x, min_y, max_x, max_y })
@@ -49,12 +49,12 @@ impl PyQuadTree {
             .collect()
     }
 
-    pub fn nearest_neighbor(&self, xy: (f64, f64)) -> Option<(u64, f64, f64)> {
+    pub fn nearest_neighbor(&self, xy: (f32, f32)) -> Option<(u64, f32, f32)> {
         let (x, y) = xy;
         self.inner.nearest_neighbor(Point { x, y }).map(item_to_tuple)
     }
 
-    pub fn nearest_neighbors(&self, xy: (f64, f64), k: usize) -> Vec<(u64, f64, f64)> {
+    pub fn nearest_neighbors(&self, xy: (f32, f32), k: usize) -> Vec<(u64, f32, f32)> {
         let (x, y) = xy;
         self.inner
             .nearest_neighbors(Point { x, y }, k)
@@ -64,7 +64,7 @@ impl PyQuadTree {
     }
 
     /// Returns all rectangle boundaries in the quadtree for visualization
-    pub fn get_all_rectangles(&self) -> Vec<(f64, f64, f64, f64)> {
+    pub fn get_all_rectangles(&self) -> Vec<(f32, f32, f32, f32)> {
         self.inner
             .get_all_rectangles()
             .into_iter()

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/src/quadtree.rs

@@ -22,6 +22,7 @@ pub struct QuadTree {
 // 1: (x >= cx, y < cy)
 // 2: (x < cx, y >= cy)
 // 3: (x >= cx, y >= cy)
+#[inline(always)]
 fn child_index_for_point(b: &Rect, p: &Point) -> usize {
     let cx = 0.5 * (b.min_x + b.max_x);
     let cy = 0.5 * (b.min_y + b.max_y);
@@ -34,7 +35,7 @@ impl QuadTree {
     pub fn new(boundary: Rect, capacity: usize) -> Self {
         QuadTree {
             boundary,
-            items: Vec::new(),
+            items: Vec::with_capacity(capacity),
             capacity,
             children: None,
             depth: 0,
@@ -45,7 +46,7 @@ impl QuadTree {
     pub fn new_with_max_depth(boundary: Rect, capacity: usize, max_depth: usize) -> Self {
         QuadTree {
             boundary,
-            items: Vec::new(),
+            items: Vec::with_capacity(capacity),
             capacity,
             children: None,
             depth: 0,
@@ -56,7 +57,7 @@ impl QuadTree {
     pub fn new_child(boundary: Rect, capacity: usize, depth: usize, max_depth: usize) -> Self {
         QuadTree {
             boundary,
-            items: Vec::new(),
+            items: Vec::with_capacity(capacity),
             capacity,
             children: None,
             depth: depth,
@@ -121,42 +122,38 @@ impl QuadTree {
 
     pub fn query(&self, range: Rect) -> Vec<Item> {
         let mut out = Vec::new();
-        self.query_into(&range, &mut out);
-        out
-    }
+        let mut stack: Vec<&QuadTree> = Vec::new();
+        stack.push(self);
 
-    fn query_into(&self, range: &Rect, out: &mut Vec<Item>) {
-        // prune if this node does not intersect the query
-        if !self.boundary.intersects(range) {
-            return;
-        }
-
-        // check items stored at this node
-        for it in &self.items {
-            if range.contains(&it.point) {
-                out.push(*it); // Item is Copy
+        while let Some(node) = stack.pop() {
+            for it in &node.items {
+                if range.contains(&it.point) {
+                    out.push(*it);
+                }
             }
-        }
-
-        // recurse to children
-        if let Some(children) = self.children.as_ref() {
-            for child in children.iter() {
-                child.query_into(range, out);
+            if let Some(children) = node.children.as_ref() {
+                // Push children that intersect the query range
+                for child in children.iter() {
+                    if range.intersects(&child.boundary) {
+                        stack.push(child);
+                    }
+                }
             }
         }
+        out
     }
 
     pub fn nearest_neighbor(&self, point: Point) -> Option<Item> {
-        self.nearest_neighbors_within(point, 1, f64::INFINITY)
+        self.nearest_neighbors_within(point, 1, f32::INFINITY)
             .into_iter()
             .next()
     }
 
     pub fn nearest_neighbors(&self, point: Point, k: usize) -> Vec<Item> {
-        self.nearest_neighbors_within(point, k, f64::INFINITY)
+        self.nearest_neighbors_within(point, k, f32::INFINITY)
     }
 
-    pub fn nearest_neighbors_within(&self, point: Point, k: usize, max_distance: f64) -> Vec<Item> {
+    pub fn nearest_neighbors_within(&self, point: Point, k: usize, max_distance: f32) -> Vec<Item> {
         if k == 0 {
             return Vec::new();
         }
@@ -167,7 +164,7 @@ impl QuadTree {
 
         for _ in 0..k {
             // stack holds (node_ref, bbox_distance_sq)
-            let mut stack: Vec<(&QuadTree, f64)> = Vec::new();
+            let mut stack: Vec<(&QuadTree, f32)> = Vec::new();
             stack.push((self, dist_sq_point_to_rect(&point, &self.boundary)));
 
             let mut best: Option<Item> = None;
@@ -181,7 +178,7 @@ impl QuadTree {
 
                 if let Some(children) = node.children.as_ref() {
                     // compute and sort children by bbox distance, push farthest first
-                    let mut kids: Vec<(&QuadTree, f64)> = children
+                    let mut kids: Vec<(&QuadTree, f32)> = children
                         .iter()
                         .map(|c| (c, dist_sq_point_to_rect(&point, &c.boundary)))
                         .filter(|(_, d2)| *d2 < best_d2)

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/tests/insertions.rs

@@ -1,10 +1,10 @@
 use fastquadtree::{Point, Rect, Item, QuadTree};
 
-fn r(x0: f64, y0: f64, x1: f64, y1: f64) -> Rect {
+fn r(x0: f32, y0: f32, x1: f32, y1: f32) -> Rect {
     Rect { min_x: x0, min_y: y0, max_x: x1, max_y: y1 }
 }
 
-fn pt(x: f64, y: f64) -> Point { Point { x, y } }
+fn pt(x: f32, y: f32) -> Point { Point { x, y } }
 
 #[test]
 fn rect_contains_half_open() {
@@ -73,7 +73,7 @@ fn many_inserts_all_succeed_when_inside() {
     let mut id = 1u64;
     for x in (10..1000).step_by(100) {
         for y in (10..1000).step_by(100) {
-            if qt.insert(Item { id, point: pt(x as f64, y as f64) }) {
+            if qt.insert(Item { id, point: pt(x as f32, y as f32) }) {
                 ok += 1;
             }
             id += 1;

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/tests/nearest_neighbor.rs

@@ -1,15 +1,15 @@
 use fastquadtree::{Point, Rect, Item, QuadTree};
 
-fn r(x0: f64, y0: f64, x1: f64, y1: f64) -> Rect {
+fn r(x0: f32, y0: f32, x1: f32, y1: f32) -> Rect {
     Rect { min_x: x0, min_y: y0, max_x: x1, max_y: y1 }
 }
-fn pt(x: f64, y: f64) -> Point { Point { x, y } }
+fn pt(x: f32, y: f32) -> Point { Point { x, y } }
 fn ids(v: &[Item]) -> Vec<u64> {
     let mut out: Vec<u64> = v.iter().map(|it| it.id).collect();
     out.sort_unstable();
     out
 }
-fn dist2(a: Point, b: Point) -> f64 {
+fn dist2(a: Point, b: Point) -> f32 {
     let dx = a.x - b.x;
     let dy = a.y - b.y;
     dx * dx + dy * dy
@@ -69,7 +69,7 @@ fn knn_basic_ordering_no_split() {
     assert_eq!(order, vec![2, 1, 4, 3, 5]);
 
     // distances should be nondecreasing
-    let d: Vec<f64> = res.iter().map(|it| dist2(q, it.point)).collect();
+    let d: Vec<f32> = res.iter().map(|it| dist2(q, it.point)).collect();
     for w in d.windows(2) {
         assert!(w[0] <= w[1] + 1e-12);
     }
@@ -176,7 +176,7 @@ fn ordering_is_by_distance_even_after_splits() {
     let mut id = 1u64;
     for y in (4..=60).step_by(8) {
         for x in (4..=60).step_by(8) {
-            qt.insert(Item { id, point: pt(x as f64, y as f64) });
+            qt.insert(Item { id, point: pt(x as f32, y as f32) });
             id += 1;
         }
     }
@@ -185,7 +185,7 @@ fn ordering_is_by_distance_even_after_splits() {
     assert!(!res.is_empty());
 
     // distances should be nondecreasing
-    let d: Vec<f64> = res.iter().map(|it| dist2(q, it.point)).collect();
+    let d: Vec<f32> = res.iter().map(|it| dist2(q, it.point)).collect();
     for w in d.windows(2) {
         assert!(w[0] <= w[1] + 1e-12);
     }

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/tests/query.rs

@@ -1,9 +1,9 @@
 use fastquadtree::{Point, Rect, Item, QuadTree};
 
-fn r(x0: f64, y0: f64, x1: f64, y1: f64) -> Rect {
+fn r(x0: f32, y0: f32, x1: f32, y1: f32) -> Rect {
     Rect { min_x: x0, min_y: y0, max_x: x1, max_y: y1 }
 }
-fn pt(x: f64, y: f64) -> Point { Point { x, y } }
+fn pt(x: f32, y: f32) -> Point { Point { x, y } }
 fn ids(v: &[Item]) -> Vec<u64> {
     let mut out: Vec<u64> = v.iter().map(|it| it.id).collect();
     out.sort_unstable();
@@ -101,8 +101,8 @@ fn query_range_covering_multiple_children_returns_union() {
 fn full_range_query_returns_all_items() {
     let mut qt = QuadTree::new(r(0.0, 0.0, 100.0, 100.0), 3);
     for i in 0..9 {
-        let x = 10.0 + 10.0 * (i as f64);
-        let y = 20.0 + 5.0 * (i as f64);
+        let x = 10.0 + 10.0 * (i as f32);
+        let y = 20.0 + 5.0 * (i as f32);
         assert!(qt.insert(Item { id: i + 1, point: pt(x, y) }));
     }
     let hits = qt.query(r(0.0, 0.0, 100.0, 100.0));

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/tests/test_wrapper_edges.py

@@ -207,3 +207,20 @@ def test_insert_many_points_exception_for_out_of_bounds():
 
     with pytest.raises(ValueError):
         qt.insert_many_points(points)
+
+
+def test_auto_id_collision_prevention():
+    qt = QuadTree(BOUNDS, capacity=8, track_objects=True)
+    id1 = qt.insert((10, 10))  # auto ID
+
+    id2 = qt.insert((20, 20), id=200)  # Large ID
+
+    # Next auto ID should be 201
+    id3 = qt.insert((30, 30))  # auto ID
+
+    assert id3 == 201
+    assert len(qt) == 3
+    assert id1 != id2 != id3
+
+    id4 = qt.insert((40, 40), id=150)  # Manual ID lower than current auto ID
+    assert id4 == 150

{fastquadtree-0.4.1 → fastquadtree-0.5.0}/tests/unconventional_bounds.rs

@@ -1,14 +1,14 @@
 use fastquadtree::{Point, Rect, Item, QuadTree};
 
-fn r(x0: f64, y0: f64, x1: f64, y1: f64) -> Rect {
+fn r(x0: f32, y0: f32, x1: f32, y1: f32) -> Rect {
     Rect { min_x: x0, min_y: y0, max_x: x1, max_y: y1 }
 }
 
-fn pt(x: f64, y: f64) -> Point { 
+fn pt(x: f32, y: f32) -> Point { 
     Point { x, y } 
 }
 
-fn item(id: u64, x: f64, y: f64) -> Item {
+fn item(id: u64, x: f32, y: f32) -> Item {
     Item { id, point: pt(x, y) }
 }
 
@@ -226,8 +226,8 @@ fn stress_test_negative_coordinates() {
     let mut inserted_count = 0;
     for i in 0..50 {
         for j in 0..50 {
-            let x = -20.0 * (i as f64) - 10.0; // Range from -10 to -990
-            let y = -20.0 * (j as f64) - 10.0; // Range from -10 to -990
+            let x = -20.0 * (i as f32) - 10.0; // Range from -10 to -990
+            let y = -20.0 * (j as f32) - 10.0; // Range from -10 to -990
             let id = (i * 50 + j) as u64 + 1;
             if x >= -1000.0 && y >= -1000.0 && x < 0.0 && y < 0.0 {
                 assert!(qt.insert(item(id, x, y)));

fastquadtree-0.4.1/examples/delete_demo.rs

@@ -1,63 +0,0 @@
-use fastquadtree::{QuadTree, Item, Point, Rect};
-
-fn main() {
-    println!("=== QuadTree Delete by ID+Location Demo ===\n");
-    
-    // Create a quadtree
-    let mut tree = QuadTree::new(Rect { min_x: 0.0, min_y: 0.0, max_x: 100.0, max_y: 100.0 }, 4);
-    
-    // Insert multiple items at the same location
-    let shared_location = Point { x: 50.0, y: 50.0 };
-    tree.insert(Item { id: 1, point: shared_location });
-    tree.insert(Item { id: 2, point: shared_location });
-    tree.insert(Item { id: 3, point: shared_location });
-    
-    // Insert items at different locations
-    tree.insert(Item { id: 4, point: Point { x: 25.0, y: 25.0 } });
-    tree.insert(Item { id: 5, point: Point { x: 75.0, y: 75.0 } });
-    
-    println!("Initial tree with {} items", tree.count_items());
-    
-    // Query the area around the shared location
-    let query_rect = Rect { min_x: 45.0, min_y: 45.0, max_x: 55.0, max_y: 55.0 };
-    let items_at_location = tree.query(query_rect);
-    println!("Items at (50,50): {:?}", items_at_location.iter().map(|i| i.id).collect::<Vec<_>>());
-    
-    // Delete specific item by ID+location - only removes that specific item
-    println!("\nDeleting item with ID=2 at (50,50)...");
-    let deleted = tree.delete(2, shared_location);
-    println!("Delete successful: {}", deleted);
-    println!("Tree now has {} items", tree.count_items());
-    
-    // Verify the other items at the same location are still there
-    let remaining_items = tree.query(query_rect);
-    println!("Remaining items at (50,50): {:?}", remaining_items.iter().map(|i| i.id).collect::<Vec<_>>());
-    
-    // Try to delete the same item again - should fail
-    println!("\nTrying to delete ID=2 again...");
-    let deleted_again = tree.delete(2, shared_location);
-    println!("Delete successful: {}", deleted_again);
-    
-    // Try to delete with wrong ID - should fail  
-    println!("\nTrying to delete ID=999 at (50,50)...");
-    let wrong_id = tree.delete(999, shared_location);
-    println!("Delete successful: {}", wrong_id);
-    
-    // Try to delete with wrong location - should fail
-    println!("\nTrying to delete ID=1 at wrong location (60,60)...");
-    let wrong_location = tree.delete(1, Point { x: 60.0, y: 60.0 });
-    println!("Delete successful: {}", wrong_location);
-    
-    // Delete the remaining items at the shared location
-    println!("\nDeleting remaining items at (50,50)...");
-    tree.delete(1, shared_location);
-    tree.delete(3, shared_location);
-    
-    println!("Final tree has {} items", tree.count_items());
-    
-    // Verify the items at other locations are still there
-    let all_items = tree.query(Rect { min_x: 0.0, min_y: 0.0, max_x: 100.0, max_y: 100.0 });
-    println!("All remaining items: {:?}", all_items.iter().map(|i| (i.id, i.point.x, i.point.y)).collect::<Vec<_>>());
-    
-    println!("\n=== Demo Complete ===");
-}