Geo Polygonize

A native Rust port of the JTS/GEOS polygonization algorithm. This crate allows you to reconstruct valid polygons from a set of lines, including handling of complex topologies like holes, nested shells, and disconnected components.

Ask DeepWiki

Features

  • Robust Polygonization: Extracts polygons from unstructured linework.
  • Iterative Grid Noding (Unchecked): Splits and snaps dirty linework, without claiming certified snap-rounding guarantees.
  • Hardware Acceleration: Uses SIMD instructions (via wide crate) for critical geometric predicates like Point-in-Polygon checks.
  • Wasm Optimized: Tailored for WebAssembly with talc allocator and binary GeoArrow support.
  • Performance: SIMD, spatial indexing, and optional parallel execution, with checked-in benchmark tooling.
  • Geo Ecosystem: Fully integrated with geo-types and geo crates.
  • GeoArrow Support: Arrow C Data Interface and Arrow IPC integration with GeoArrow metadata.

Engineering Roadmap

For an ambitious, prioritized plan covering performance, security, API consistency, and maintainability, see ROADMAP.md.

Usage

Library

use geo_polygonize_core::{polygonize, Coord3D, Line3D};
use geo_polygonize_core::options::PolygonizerOptions;

fn main() {
    let points = [
        Coord3D::new(0.0, 0.0, 0.0),
        Coord3D::new(10.0, 0.0, 0.0),
        Coord3D::new(10.0, 10.0, 0.0),
        Coord3D::new(0.0, 10.0, 0.0),
    ];
    let lines = (0..4).map(|i| Line3D::new(points[i], points[(i + 1) % 4], i as u32));

    let result = polygonize(lines, &PolygonizerOptions::default())
        .expect("Polygonization failed");

    for polygon in result.polygons {
        println!("Found polygon with area: {}", polygon.unsigned_area_2d());
    }
}

Choosing node_input and snap_grid_size

Polygonization quality is heavily influenced by input noding strategy.

  • node_input = false (default): Fastest path. Use this when your input linework is already noded (all intersections are explicit vertices).
  • node_input = true: Enables unchecked iterative grid noding. Use this for real-world datasets that may contain slight misalignments, overlaps, or self-intersections, and validate outputs when correctness must be certified.
  • snap_grid_size controls how aggressively coordinates are snapped during robust noding:
    • Start with 1e-10 for high-precision projected data.
    • Increase to 1e-8 or 1e-6 when near-duplicate vertices prevent clean topology.
    • Avoid very large values unless your coordinate units are coarse; oversnapping can collapse narrow features.

Practical workflow:

  1. Run with node_input = false first on trusted data.
  2. If you observe missing polygons, sliver artifacts, or unresolved intersections, enable node_input.
  3. Tune snap_grid_size upward incrementally until topology stabilizes.

Output semantics

The polygonizer intentionally returns only valid polygonal areas that can be formed from closed cycles:

  • Dangles are removed: dead-end edges do not appear in output polygons.
  • Cut edges are excluded: edges that are connected but cannot bound a face are ignored.
  • Holes and nested shells are preserved when enough boundary information is present.

This behavior matches classical JTS/GEOS polygonization semantics and is useful for cleaning linework before area analysis.

GeoArrow Integration

The library supports ingesting data directly from Arrow arrays via the arrow_api module and ffi.

use geo_polygonize_core::arrow_api::{polygonize_arrow, PolygonizerOptions};
// ... create Arrow array ...
// let result = polygonize_arrow(&array, &field, options);

Python

The Python package is published as geo-polygonize-py and imported as geo_polygonize.

pip install geo-polygonize-py

import numpy as np
from geo_polygonize import polygonize, import_probe

# 1. Using Shapely LineStrings or coordinate lists directly
lines = [
    [(0, 0), (10, 0), (10, 10), (0, 10), (0, 0)],
    [(0, 0), (10, 10)]
]

# return_polygons=True returns a list of shapely.geometry.Polygon objects
polygons = polygonize(lines=lines, return_polygons=True)
for p in polygons:
    print(p.area)

# 2. Using High-Performance Flat Arrays
# Flat buffers avoid Python object-per-coordinate overhead
coords = np.array([
    0.0, 0.0, 10.0, 0.0, 10.0, 10.0, 0.0, 10.0, 0.0, 0.0,
    0.0, 0.0, 10.0, 10.0
], dtype=np.float64)

# Start indices for each line segment.
# The final closing offset is computed implicitly.
offsets = np.array([0, 5], dtype=np.uint32)

# Returns a stable dictionary with 'polygons', diagnostics, and provenance.
result_dict = polygonize(coords=coords, offsets=offsets)

# Native-extension probes are cheap and safe for optional integrations.
ok, error = import_probe()

CFB/autograder integrations should use the versioned production profile rather than assembling caller-side knobs or using legacy polygonize(..., node=True, snap=0.5) calls:

from geo_polygonize import cfb_robust_options, polygonize_with_options

result = polygonize_with_options(
    coords=coords,
    offsets=offsets,
    options=cfb_robust_options(),
)

The default return shape is a stable dictionary with polygons as SimplePolygon values. Use return_polygons=True only when you want Shapely Polygon objects.

SnapStrategy::Grid keeps topology and output coordinates on the configured precision grid. The CFB profile uses GeosCompat: the grid establishes robust topology, then output nodes regain deterministic source coordinates to better match Shapely snap plus full-precision noding. It is not set_precision emulation, and exact parity is not guaranteed for many-to-one snaps.

For Shapely parity checks, compare report-mode outputs with the built-in mismatch helper:

from geo_polygonize import explain_mismatch, polygonize_with_options

options = cfb_robust_options()
result_a = polygonize_with_options(coords=coords_a, offsets=offsets_a, options=options)
result_b = polygonize_with_options(coords=coords_b, offsets=offsets_b, options=options)
result_a["options"] = options
result_b["options"] = options

mismatch = explain_mismatch(result_a, result_b)

For a minimal Shapely smoke comparison, use area signatures:

from shapely.ops import polygonize as shapely_polygonize

rust_polys = polygonize_with_options(lines=lines, options=cfb_robust_options(), return_polygons=True)
rust_areas = sorted(round(poly.area, 6) for poly in rust_polys)
shapely_areas = sorted(round(poly.area, 6) for poly in shapely_polygonize(lines))

WebAssembly (WASM)

This library supports WebAssembly with an ergonomic dual-build configuration that automatically utilizes SIMD instructions where available.

Installation:

npm install geo-polygonize

Standard Usage (Quick Demos): The default entry point automatically handles feature detection (SIMD) and lazy-loading of the Wasm binary. The Wasm is inlined as a Base64 Data URI, so no extra bundler configuration is needed. For app builds, prefer the slim entry point below so your bundler keeps the Wasm assets out of the JavaScript chunk.

import init, { polygonize, polygonize_geoarrow } from "geo-polygonize";

async function run() {
    await init();

    const geojson = {
        "type": "FeatureCollection",
        "features": [
            // ... your line features
        ]
    };

    // Returns a GeoJSON FeatureCollection string
    // Pass explicitly matching backend configuration if desired
    const result = polygonize(
        JSON.stringify(geojson),
        true, // node_input
        0.5   // snap_grid_size
    );
    console.log(JSON.parse(result));

    // Or use Arrow IPC bytes
    // const ipcBuffer = ...;
    // const arrowResult = polygonize_geoarrow(ipcBuffer, false, 1e-10, false);
}

Slim Usage (Apps / Manual Loading): For Vite and other app bundlers, import from geo-polygonize/slim and pass explicit Wasm asset URLs.

import { cfbRobustOptions, initBest } from "geo-polygonize/slim";
import scalarUrl from "geo-polygonize/geo_polygonize.wasm?url";
import simdUrl from "geo-polygonize/geo_polygonize_simd.wasm?url";

async function run() {
    const wasm = await initBest(
        { module_or_path: scalarUrl },
        { module_or_path: simdUrl },
    );

    const result = wasm.polygonizeWithOptions(
        JSON.stringify(geojson),
        cfbRobustOptions,
    );
}

Multithreaded Usage (Experimental): This library provides a multithreaded build powered by wasm-bindgen-rayon.

import init, { initThreadPool, polygonize } from "geo-polygonize/threads";

async function run() {
    await init();

    // Initialize thread pool (e.g., with navigator.hardwareConcurrency)
    await initThreadPool(navigator.hardwareConcurrency);

    // ... use polygonize as usual
}

Important: Multithreaded WebAssembly requires SharedArrayBuffer, which is only available in secure contexts. You must serve your page with the following headers:

Cross-Origin-Opener-Policy: same-origin
Cross-Origin-Embedder-Policy: require-corp

CLI Example

The repository includes a CLI tool to polygonize GeoJSON files.

# Build the example
cargo build -p geo-polygonize-core --example polygonize --release

# Run on input lines
cargo run -p geo-polygonize-core --release --example polygonize -- --input lines.geojson --output polygons.geojson --node

Visualization

You can visualize the results using the provided Python script (requires matplotlib and shapely).

python3 scripts/visualize.py --input lines.geojson --output polygons.geojson --save result.png

Examples

Below are some examples of what the polygonizer can do.

Nested Holes and Islands

The algorithm correctly identifies nested structures (Island inside a Hole inside a Shell).

Nested Holes

Incomplete Grid / Dangles

The algorithm prunes dangles (dead-end lines) and extracts only closed cycles.

Incomplete Grid

Touching Polygons (Shared Edges)

Using robust noding (--node), it can reconstruct adjacent polygons that share boundaries, even if the input lines are not perfectly noded.

Touching Polygons

Self-Intersecting Geometry (Bowtie)

Self-intersecting lines are split at intersection points, and valid cycles are extracted.

Bowtie

Complex Geometries

The polygonizer can handle complex, curved inputs (approximated by LineStrings) such as overlapping circles and shapes with multiple holes.

Overlapping Circles: Note how the intersection regions are correctly identified as separate polygons.

Overlapping Circles

Curved Holes: A complex polygon with multiple circular holes.

Curved Holes

Benchmarks

This library includes a "severe" comparison suite against shapely (GEOS).

See BENCHMARKS.md for detailed results and instructions on how to run them.

Architecture

This implementation moves away from the pointer-based graph structures of JTS/GEOS to a Rust-idiomatic Index Graph (Arena) approach.

See ARCHITECTURE.md for a deep dive into the optimization strategies.

Key optimizations include:

  1. Noding: Unchecked iterative grid noding with spatially dispatched intersection detection.
  2. Vectorization: SIMD-accelerated Ray Casting for efficient Hole Assignment.
  3. Memory Layout: Structure of Arrays (SoA) for graph nodes and talc allocator for Wasm.

License

MIT/Apache-2.0

Key dates

PyPI data
Data sourced directly from PyPI's database.
  • Released:
Newer release available (1.1.0)

1 maintainer

PyPI data
Data sourced directly from PyPI's database.
Avatar for graydonpleasants from gravatar.com graydonpleasants

License

MIT OR Apache-2.0

Requires

Python >=3.8

Classifiers

Report project as malware

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

geo_polygonize_py-0.40.0.tar.gz (145.5 kB view details)

Uploaded Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

geo_polygonize_py-0.40.0-cp38-abi3-win_amd64.whl (875.0 kB view details)

Uploaded CPython 3.8+Windows x86-64

geo_polygonize_py-0.40.0-cp38-abi3-manylinux_2_35_x86_64.whl (1.0 MB view details)

Uploaded CPython 3.8+manylinux: glibc 2.35+ x86-64

geo_polygonize_py-0.40.0-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl (961.4 kB view details)

Uploaded CPython 3.8+manylinux: glibc 2.17+ ARM64

geo_polygonize_py-0.40.0-cp38-abi3-macosx_11_0_arm64.whl (872.0 kB view details)

Uploaded CPython 3.8+macOS 11.0+ ARM64

File details

Details for the file geo_polygonize_py-0.40.0.tar.gz.

File metadata

  • Download URL: geo_polygonize_py-0.40.0.tar.gz
  • Upload date:
  • Size: 145.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.14

File hashes

Hashes for geo_polygonize_py-0.40.0.tar.gz
Algorithm Hash digest
SHA256 a1e659ab3266a6838a6ae616021095f22fc24a0e5bde4f6b455fa59f4fecc951
MD5 ed2a9a1e7fe754e179b0c111027f7d0c
BLAKE2b-256 a3aba38fc60bc185a361c507d55e75fa49340cc400fa4d84a0b470e828300412

See more details on using hashes here.

Provenance

The following attestation bundles were made for geo_polygonize_py-0.40.0.tar.gz:

Publisher: publish-python.yml on graydonpleasants/geo-polygonize

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file geo_polygonize_py-0.40.0-cp38-abi3-win_amd64.whl.

File metadata

File hashes

Hashes for geo_polygonize_py-0.40.0-cp38-abi3-win_amd64.whl
Algorithm Hash digest
SHA256 052ec406444c556a00a88b427714ebe9da9d07d58b879020aeca27bbb0ac9812
MD5 d3b4ff377baccfeb31245c3744a48d30
BLAKE2b-256 bfdd6379ed6f905ddb9259ef7ecc480d608ef9d79acbd820aa52220b3894a318

See more details on using hashes here.

Provenance

The following attestation bundles were made for geo_polygonize_py-0.40.0-cp38-abi3-win_amd64.whl:

Publisher: publish-python.yml on graydonpleasants/geo-polygonize

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file geo_polygonize_py-0.40.0-cp38-abi3-manylinux_2_35_x86_64.whl.

File metadata

File hashes

Hashes for geo_polygonize_py-0.40.0-cp38-abi3-manylinux_2_35_x86_64.whl
Algorithm Hash digest
SHA256 01e3e49aa0d81ce91801bd827e5154e1eaf30804bb088c410eb83ddfc0e1fea8
MD5 16feffa4cca9cafad69da5275fd13c7f
BLAKE2b-256 36fddb8407ef6fc4ac07d70152e56248faa9e275254c29e929bc1b288de83505

See more details on using hashes here.

Provenance

The following attestation bundles were made for geo_polygonize_py-0.40.0-cp38-abi3-manylinux_2_35_x86_64.whl:

Publisher: publish-python.yml on graydonpleasants/geo-polygonize

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file geo_polygonize_py-0.40.0-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl.

File metadata

File hashes

Hashes for geo_polygonize_py-0.40.0-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
Algorithm Hash digest
SHA256 0427938ad878029a71c4b7eb3ccf73f14ba85787a6d8bbaea2f967d122a7d077
MD5 09b6bb90ce57160f2112621c7c07c972
BLAKE2b-256 3f5c55d9819df586bc4ee4712599fcd44bdde86dcb7ff5cd7a14f490f940ee72

See more details on using hashes here.

Provenance

The following attestation bundles were made for geo_polygonize_py-0.40.0-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl:

Publisher: publish-python.yml on graydonpleasants/geo-polygonize

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file geo_polygonize_py-0.40.0-cp38-abi3-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for geo_polygonize_py-0.40.0-cp38-abi3-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 3bd07921e23767477571bb76a634c9a365cab92044811fc6166f7fc5decd0c15
MD5 59416416525fa60a3e233bbdb1a5b288
BLAKE2b-256 98c3840f0865fb10e58a260c4c8097c12f8469244e1156f98a964c2f87accfad

See more details on using hashes here.

Provenance

The following attestation bundles were made for geo_polygonize_py-0.40.0-cp38-abi3-macosx_11_0_arm64.whl:

Publisher: publish-python.yml on graydonpleasants/geo-polygonize

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

1.1.0

5 files

1.0.0

5 files

0.76.2

5 files

0.76.1

5 files

0.76.0

5 files

0.75.0

5 files

0.74.0

5 files

0.73.0

5 files

0.72.0

5 files

0.71.0

5 files

0.70.0

5 files

0.69.0

5 files

0.68.0

5 files

0.67.0

5 files

0.66.0

5 files

0.65.0

5 files

0.64.0

5 files

0.63.0

5 files

0.62.0

5 files

0.61.0

5 files

0.60.0

5 files

0.59.0

5 files

0.58.1

5 files

0.58.0

5 files

0.57.0

5 files

0.56.0

5 files

0.55.0

5 files

0.54.0

5 files

0.53.0

5 files

0.52.0

5 files

0.51.2

5 files

0.51.1

5 files

0.51.0

5 files

0.50.0

5 files

0.49.0

5 files

0.48.0

5 files

0.47.1

5 files

0.47.0

5 files

0.46.2

5 files

0.46.1

5 files

0.46.0

5 files

0.45.1

5 files

0.45.0

5 files

0.44.0

5 files

0.43.0

5 files

0.42.0

5 files

0.41.0

5 files

0.40.1

5 files

This release

0.40.0 This release

5 files

0.39.13

5 files

0.39.12

5 files

0.39.11

5 files

0.39.10

5 files

0.39.9

5 files

0.39.8

5 files

0.39.7

5 files

0.39.6

5 files

0.39.5

5 files

0.39.4

5 files

0.39.3

5 files

0.39.2

5 files

0.39.1

5 files

0.39.0

5 files

0.38.1

5 files

0.38.0

5 files

0.37.8

5 files

0.37.7

5 files

0.37.6

5 files

0.37.5

5 files

0.37.4

5 files

0.37.3

5 files

0.37.2

5 files

0.37.1

5 files

0.37.0

5 files

0.36.2

5 files

0.36.1

5 files

0.36.0

5 files

0.35.2

5 files

0.35.1

4 files

0.35.0

4 files

0.34.0

3 files

0.33.1

3 files

0.33.0

3 files

0.23.1

3 files

0.23.0

3 files

0.22.1

3 files

0.22.0

3 files

0.21.0

3 files

0.20.0

3 files

0.19.0

3 files

0.18.1

3 files

0.18.0

3 files

0.17.4

3 files

0.17.3

3 files

0.17.2

3 files

0.17.1

3 files

0.17.0

3 files

0.16.0

3 files

0.15.0

3 files

0.14.1

3 files

0.14.0

3 files

0.13.0

3 files

0.12.1

3 files

0.12.0

3 files

0.11.0

3 files

0.10.0

3 files

0.9.0

3 files

0.8.1

3 files

0.8.0

3 files

0.7.0

3 files

0.6.3

3 files

0.6.2

3 files

0.6.1

3 files

0.6.0

3 files

0.5.0

3 files

0.4.2

3 files

0.4.1

3 files

Yanked

0.1.0

5 files

Yanked reason: wrong version number

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page