pypgl

⚠️ Work in Progress: This library is still under construction and contains bugs and missing features. Use in production environments is not recommended.

Pangolin (or pgl) is a C++ library for computational geometry in the plane and pypgl is the official python binding for it. It is designed to be pleasant to use and always exact. Calculations are exact using rational numbers (floating point is not accepted).

import pypgl as pgl

p = pgl.Point(1, 0)
q = pgl.Point(4, '15/2')
s = pgl.Segment(p, q)
t = pgl.Segment(0, 8, '7/3', 1)
if s.intersects(t):
    print(s, "intersects", t)
# Output: (1,0)--(4,15/2) intersects (0,8)--(7/3,1)

There are many more illustrated examples that give a good overview of the library's features and syntax.

Shapes and Predicates

Family	Shapes
0-dimensional	`Point`
1-dimensional	`Segment`, `OrientedSegment`, `Line`, `OrientedLine`, `Ray`, `Polyline`, `MonotoneChain`
2-dimensional	`Halfplane`, `Triangle`, `Rectangle`, `Disk`, `Convex`, `Polygon`, `PolygonWithHoles`, `PolygonSet`, `HalfplaneIntersection`

The following predicates are implemented as methods of all shapes.

contains(Shape) Does it contain the other shape?
boundaryContains(Shape) Does its boundary contain the other shape?
interiorContains(Shape) Does it contain the other shape in the interior?
intersects(Shape) Do the two shapes intersect?
interiorsIntersect(Shape) Do the interiors of the two shapes intersect?
separates(Shape) Does one shape cut the other into two (or more) components?
crosses(Shape) Do both shapes separate each other?

import pypgl as pgl

o = pgl.Point()      # Point (0,0)
d = pgl.Disk(o, 10)  # Disk of radius 10 centered at (0,0)
if d.contains(o):
    print("Disk contains", o)
diam = d.diameter()
if d.contains(diam):
    print("Disk contains the diameter")
if not d.interiorContains(diam):
    print("Disk's interior does not contain the diameter")
# Output:
# Disk contains (0,0)
# Disk contains the diameter
# Disk's interior does not contain the diameter

Other Methods

Several other methods are supported by the shapes.

import pypgl as pgl

c = pgl.Convex([pgl.Point(0, 0), pgl.Point(1, 0), pgl.Point(1, 2), pgl.Point(0, 1)])
s = c.diameter()
print("The diameter of", c,
      "is defined by", s,
      "and has length", s.length())
# Output: The diameter of Convex[(0,0),(1,0),(1,2),(0,1)] is defined by (0,0)--(1,2) and has length 2.23607

Distances come in the Euclidean (squaredDistance, exact and therefore squared), Manhattan (distanceL1) and Chebyshev (distanceLInf) flavors, each with a Hausdorff variant between shapes. samePointSet asks whether two shapes cover the same points, across types and regardless of how each is written.

The boolean operations (regularizedUnion, difference, symmetricDifference, regularizedIntersection) and the Minkowski sum are closed over the region shapes, so a result feeds straight back in. Its dual, the Minkowski erosion, answers where a shape fits inside another:

import pypgl as pgl

square = pgl.Polygon([pgl.Point(0,0), pgl.Point(10,0), pgl.Point(10,10), pgl.Point(0,10)])
holed = square.difference(pgl.Rectangle(pgl.Point(3,3), pgl.Point(7,7)))
print(holed.area(), holed.holeCount())
# Output: 84 1

Shapes are moved around with +/-/*//, and an arbitrary affine map is applied with a Transformation:

import pypgl as pgl

t = pgl.Transformation.rotation90() * pgl.Transformation.translation(2, 0)
print(t * pgl.Segment(0, 0, 5, 5))
# Output: (-5,7)--(0,2)

Visualization

A Canvas class is provided for visualization, exporting to SVG, PDF, or Ipe:

import pypgl as pgl

canvas = pgl.Canvas()
canvas.draw(pgl.Point(0, 0))

tri = pgl.Triangle(-1, -1, 0, 2, 1, -2)
canvas.stroke("green")
canvas.draw(tri)
canvas.stroke("blue")
canvas.draw(2*tri)
canvas.writeSVG("example2.svg")

Algorithms and Data Structures

Pangolin includes fundamental algorithms:

Convex hull computed with Graham scan.
Line segment intersection: Bentley-Ottmann sweep line using rational numbers.
Minkowski sum, Minkowski erosion and boolean operations.
Visibility graph and visibility polygon.
Find the closest pair of points using divide and conquer.
Smallest enclosing disk and rectangle.
Sort points by angle or Hilbert order.

and data structures:

Kd-tree for points and a generalization for other bounded shapes.
Interval tree to use 1-dimensional queries on projections.
Triangulation including Delaunay and constrained Delaunay triangulations for points and polygons.
Arrangement of lines, line segments, and rays with a trapezoidal map for fast point location.
Graph class for combinatorial algorithms like Dijkstra and Prim that can be used to compute Euclidean minimum spanning trees and shortest paths among obstacles.

Installation

pypgl requires Python 3.10 or newer. Up to 0.6.0 the floor was 3.9, which reached end-of-life in October 2025; 3.9 users can stay on that release.

From PyPI

pip install pypgl

Pre-built wheels are published for CPython 3.10–3.14 on Linux (manylinux_2_28, x86_64), macOS (Apple Silicon), and Windows, so most users need no compiler.

From source

Installing from a source tree or directly from GitHub builds the extension locally and therefore needs a C++20 compiler (GCC 12+, Clang 15+, or, on Windows, the LLVM/ClangCL toolset). The header-only pgl library is fetched automatically by CMake — nothing else to install.

pip install git+https://github.com/gfonsecabr/pypgl.git

Development install

Work on the bindings from a checkout with an editable, in-place build:

git clone https://github.com/gfonsecabr/pypgl.git
cd pypgl
python3 -m venv .venv
.venv/bin/pip install scikit-build-core nanobind pytest
.venv/bin/pip install -e . --no-build-isolation
.venv/bin/python -m pytest tests/ -q

--no-build-isolation lets CMake find the venv's nanobind. Re-run the pip install -e . step after editing any src/*.cpp, since the editable install is what rebuilds the extension. To build against a local pgl checkout instead of the pinned upstream commit:

.venv/bin/pip install -e . --no-build-isolation \
  -C cmake.define.PGL_INCLUDE_DIR=/path/to/pgl/include

More Information

For a brief description, check the documents at the doc folder.
Runnable scripts live in the examples folder — predicates, canvas styling, a shape gallery, ShapeTree queries, constrained triangulation, Minkowski sums, enclosing shapes, minimum spanning trees, visibility, robot motion planning, arrangements and Voronoi diagrams.
Shapes and canvases render inline in a Jupyter notebook — see canvas.md.
Check the C++ version.

Download files

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

Source Distribution

pypgl-0.7.0.tar.gz (1.9 MB view details)

Uploaded Aug 24, 2026 Source

Built Distributions

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

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

pypgl-0.7.0-cp314-cp314-win_amd64.whl (2.5 MB view details)

Uploaded Aug 24, 2026 CPython 3.14Windows x86-64

pypgl-0.7.0-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (2.1 MB view details)

Uploaded Aug 24, 2026 CPython 3.14manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

pypgl-0.7.0-cp314-cp314-macosx_11_0_arm64.whl (2.3 MB view details)

Uploaded Aug 24, 2026 CPython 3.14macOS 11.0+ ARM64

pypgl-0.7.0-cp313-cp313-win_amd64.whl (2.4 MB view details)

Uploaded Aug 24, 2026 CPython 3.13Windows x86-64

pypgl-0.7.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (2.1 MB view details)

Uploaded Aug 24, 2026 CPython 3.13manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

pypgl-0.7.0-cp313-cp313-macosx_11_0_arm64.whl (2.3 MB view details)

Uploaded Aug 24, 2026 CPython 3.13macOS 11.0+ ARM64

pypgl-0.7.0-cp312-cp312-win_amd64.whl (2.4 MB view details)

Uploaded Aug 24, 2026 CPython 3.12Windows x86-64

pypgl-0.7.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (2.1 MB view details)

Uploaded Aug 24, 2026 CPython 3.12manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

pypgl-0.7.0-cp312-cp312-macosx_11_0_arm64.whl (2.3 MB view details)

Uploaded Aug 24, 2026 CPython 3.12macOS 11.0+ ARM64

pypgl-0.7.0-cp311-cp311-win_amd64.whl (2.4 MB view details)

Uploaded Aug 24, 2026 CPython 3.11Windows x86-64

pypgl-0.7.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (2.1 MB view details)

Uploaded Aug 24, 2026 CPython 3.11manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

pypgl-0.7.0-cp311-cp311-macosx_11_0_arm64.whl (2.3 MB view details)

Uploaded Aug 24, 2026 CPython 3.11macOS 11.0+ ARM64

pypgl-0.7.0-cp310-cp310-win_amd64.whl (2.4 MB view details)

Uploaded Aug 24, 2026 CPython 3.10Windows x86-64

pypgl-0.7.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (2.1 MB view details)

Uploaded Aug 24, 2026 CPython 3.10manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

pypgl-0.7.0-cp310-cp310-macosx_11_0_arm64.whl (2.3 MB view details)

Uploaded Aug 24, 2026 CPython 3.10macOS 11.0+ ARM64

File details

Details for the file pypgl-0.7.0.tar.gz.

File metadata

Download URL: pypgl-0.7.0.tar.gz
Upload date: Aug 24, 2026
Size: 1.9 MB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for pypgl-0.7.0.tar.gz
Algorithm	Hash digest
SHA256	`f51676d79e8b27aec5ab1dc2595778585ce0989a7ab3d4d962ddd7f73d808d9b`
MD5	`6c2d10e63d29f832f05a89e6c2b6695f`
BLAKE2b-256	`a9cb188f2042b2d6bd5788af3b3e5eaa17318af32d92f1ef2ec3c8719b47f848`

Algorithm	Hash digest
SHA256	`26fc7afbd10be704ae63f12509f1f508b0c9e7518e19d17d464c4d036142bb00`
MD5	`a70b74b44b034f49ff0d4a67a956c815`
BLAKE2b-256	`22a6ac808cf63b55e4c4ac4d70feda994800dfec4e9072e243901a7ea17fd6e2`

Algorithm	Hash digest
SHA256	`6a9c8fb054fb29ddb3c0ae250ed157a7e2d50a871c05ab18c1595fe48617d388`
MD5	`cec818e78cea35e828e4d460f558253d`
BLAKE2b-256	`8ea4035cf421eca79e770d028e1e903fe52fb01809027559aedc3d6cdbdb0e6d`

Algorithm	Hash digest
SHA256	`aeab84d1a3d7cac4ec777491820e58afe2038c6afa9d2a35285f0db6d479c3de`
MD5	`99db9573df0857bc24cfe176a5cd0fd0`
BLAKE2b-256	`926272c6516bd60f7c21994d686a8965b8e1474c18d7c72af7a5c27c3de32310`

Algorithm	Hash digest
SHA256	`8db4816db265d1ae4558dc1be628010897f3f874cc30e7532e341b57f1842c0c`
MD5	`0d7c3c6f0cb12ee96165bcd10656c60b`
BLAKE2b-256	`83df418d28a5d94190479900618887ea55e985e72ff677ac9a2d5f219219df79`

Algorithm	Hash digest
SHA256	`690906a7079ef30e09683b72534d8159675b607b3d4ed5804e37de71a5c7de79`
MD5	`ca47c6f587a1a4e7ecfb5d6e9d2abe64`
BLAKE2b-256	`0f40408958e2730a218a85745b69592e7ce1bb31d96f08b46a7b9f6ed12708cd`

Algorithm	Hash digest
SHA256	`4c752011854e105a2387732d325c2c6842723de2da0d8eac26668bcbbbbcc620`
MD5	`49b075ced3fb7925cc66a9633d9d5071`
BLAKE2b-256	`0f22a30e4e9d1792fcb272277d579d4bd7c1c00b8dce84846de44836f99c04f2`

Algorithm	Hash digest
SHA256	`32ffdaa2d3f61ee17e84798205267dd7fc21aa097b195161bd425f0f68fc8e8b`
MD5	`3aca7bdf551ba2eb6fbc006c10acd5d7`
BLAKE2b-256	`c6791244d567b773e724ca00682fbab55ac32fdfaccafdb44ddd75003aeff9d0`

Algorithm	Hash digest
SHA256	`5d665c8f098ed2143cec702d2e99bc4ea9601bdeac59240729ed786c7345cad7`
MD5	`c15fc766bbd40992b12575ac772664a4`
BLAKE2b-256	`eb04cfe646f2cccae3aa438f59b6e319ac463173f56bfaf201a951daf269bc1c`

Algorithm	Hash digest
SHA256	`6d8123ed1a382755d9709538de58e6c694e125b61f41cbb99c99f8f5a5211699`
MD5	`ddd148d549988585f9a956bef8ff49b5`
BLAKE2b-256	`368cbeae5a187eec5b897ad225f36595d5b5c0bbcddc354a5d4faea576fccb4b`

Algorithm	Hash digest
SHA256	`7efba18222e1d8753156c1ffc4a3313174573f17f727dc4a343f45febdb55c79`
MD5	`7aba352dd149ea4a7fd6de619980cbc4`
BLAKE2b-256	`4c761df22ddf3c7276394026eb091872f240031af709a096aeca88ee5e144260`

Algorithm	Hash digest
SHA256	`57fa1f5eaeef7d90e743c997da61139eac490c6b7c352b93ecd0bb89ca5c6848`
MD5	`bcfd86e01061a28b8b04b7a216553e2d`
BLAKE2b-256	`17b436d34a77b0d99b2dc31e703b37b7401918b4d5c67d0b91eb32dffe45e8bf`

Algorithm	Hash digest
SHA256	`ca022e2919f650715c38a3bfd37d6ca19ed1efc37d421c22a83a8912945b8356`
MD5	`02e5059a04b313d25a0906bbce37f834`
BLAKE2b-256	`ce03de2860ed915c30e37c7e14e61875f8c7d715b3fd7b5acab0186d07918b34`

Algorithm	Hash digest
SHA256	`e50f856b4c7b4f4201e5b0abe2ab1e9691b47b8b6cd49fb84d39492be0ed7227`
MD5	`99e68c0216db195365a26dbab6ecd559`
BLAKE2b-256	`84e853bf3cc0278c434ae34755d846919306a5c835851132d5eecf4d9426a274`

Algorithm	Hash digest
SHA256	`a7c5492be5691e4303ffdae3198535df256aea738779cb5faee5daf045c6c622`
MD5	`cca4c490ca1366d798ff979469078ad9`
BLAKE2b-256	`56b73e6b6fa78b6cd715c354a2d45fd83a1e0bec3e8d25791ac389af0d475359`

Algorithm	Hash digest
SHA256	`7ff4ede8d46d9090545befe088dcbefff5ac64be0b053535e6f9cc0f639226ea`
MD5	`1ae11b5e1ecbab233e72d766e8652c26`
BLAKE2b-256	`0e9675e834887f7ba328a58a09a6d837014b9fb9bb887a5df107d472f0ffb686`

pypgl 0.7.0

Shapes and Predicates

Other Methods

Visualization

Algorithms and Data Structures

Installation

From PyPI

From source

Development install

More Information

Project links

Key dates

1 maintainer

Credits

GitHub Statistics

License

Requires

Classifiers

Download files

Source Distribution

Built Distributions

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes