Skip to main content

Parametric 2D/3D CAD modeling in pure Python — fluent API, boolean operations, built on Shapely and trimesh

Project description

ScadPy

PyPI CI coverage doc coverage

Programmatic CAD in Pure Python.Documentation

ScadPy is a parametric modeling library for Python. Define 2D shapes and 3D solids with a fluent, chainable API: boolean operations, extrusions, fillets, patterns and topology queries. Export to STL, DXF, SVG or render interactively.

Built on Shapely for 2D geometry, trimesh for 3D meshes and NumPy for topology queries (vertex coordinates, edge normals, ring types and more). If you know OpenSCAD, the approach will feel familiar.

Installation

pip install scadpy

Requirements: Python ≥ 3.12.

Quick examples

# 2D — chamfered mounting plate
from scadpy import rectangle, circle, text
import numpy as np

PLATE_WIDTH   = 80
PLATE_HEIGHT  = 50
PLATE_THICKNESS = 10
HOLE_RADIUS   = 4
HOLE_MARGIN   = 10
CHAMFER_SIZE  = 8

base  = rectangle([PLATE_WIDTH, PLATE_HEIGHT])
plate = base.chamfer(CHAMFER_SIZE)

for position, normal in zip(base.vertex_coordinates, base.vertex_normals):
    hole_center = position - HOLE_MARGIN * np.sqrt(2) * normal
    plate -= circle(HOLE_RADIUS).translate(hole_center)

plate.to_screen()
chamfered mounting plate
# 3D — extruded mounting plate with label (continues from above)
TEXT_THICKNESS = 2

extruded_plate = plate.linear_extrude(PLATE_THICKNESS)
label = text("ScadPy", size=15).linear_extrude(TEXT_THICKNESS)
extruded_plate |= label.translate(z(PLATE_THICKNESS))
extruded_plate.to_screen()
chamfered mounting plate
# 3D — parametric ball bearing
from scadpy import circle, rectangle, sphere, x, y, GRAY, ORANGE

BALL_RADIUS    = 3
RACE_RADIUS    = 15
NB_BALLS       = 11
CLEARANCE      = 0.1
RING_HEIGHT    = 7
RACE_THICKNESS = 10

groove  = circle(BALL_RADIUS + CLEARANCE) | rectangle([BALL_RADIUS, RING_HEIGHT])
race    = rectangle([RACE_THICKNESS, RING_HEIGHT]) - groove
bearing = race.radial_extrude(axis=y(), pivot=x(RACE_RADIUS)).color(GRAY)
ball    = sphere(BALL_RADIUS).color(ORANGE)
bearing += ball.radial_pattern(count=NB_BALLS, axis=y(), pivot=x(RACE_RADIUS))

bearing.to_screen()
parametric ball bearing
# 3D — dice
from scadpy import cuboid, sphere, x, y, z

SIZE = 20
dice = cuboid(SIZE)
pip  = sphere(SIZE / 12).translate(z(SIZE / 2))

one   = pip
two   = pip.translate([SIZE / 4, SIZE / 4, 0]) + pip.translate([-SIZE / 4, -SIZE / 4, 0])
three = one + two
four  = two + two.rotate(90, z())
five  = one + four
six   = four + pip.translate(x(SIZE / 4)) + pip.translate(x(-SIZE / 4))

dice -= (
    one
    + two.rotate(90, x())
    + three.rotate(90, y())
    + four.rotate(-90, y())
    + five.rotate(-90, x())
    + six.rotate(-180, x())
)

dice.to_screen()
chamfered mounting plate
# 3D — storage box
from scadpy import square, x, z

SIZE_OUTER      = 20
SIZE_INNER      = 18
FILLET          = 1
BASE_HEIGHT     = 10
CUT_HEIGHT      = 8
CAP_HEIGHT_OUTER = 1.5
CAP_HEIGHT_INNER = 3
CAP_OFFSET_X    = 25
CUT_OFFSET_Z    = 2

outer_base = square(SIZE_OUTER).fillet(FILLET).linear_extrude(BASE_HEIGHT)
inner_cut  = square(SIZE_INNER).linear_extrude(CUT_HEIGHT).translate(z(CUT_OFFSET_Z))
base       = outer_base - inner_cut

cap_outer = square(SIZE_OUTER).fillet(FILLET).linear_extrude(CAP_HEIGHT_OUTER)
cap_inner = square(SIZE_INNER).linear_extrude(CAP_HEIGHT_INNER)
cap       = (cap_outer | cap_inner).translate(x(CAP_OFFSET_X))

storage_box = base + cap
storage_box.to_screen()
chamfered mounting plate

AI integration

ScadPy ships a machine-readable skill file (ai-skills.txt) that lets AI assistants understand the full API without reading source code — signatures, descriptions, parameters, return types, and usage examples, extracted directly from the source.

Cheat sheet

Parameters shown in # comments are optional, with their default values.

2D — Shape

from scadpy import *

# primitives
circle(radius=3)                                # segment_count=64
polygon(points=[(-2, -2), (2, -2), (0, 2)])
rectangle(size=[6, 3])
Shape.from_dxf("file.dxf")
Shape.from_svg("file.svg")
square(size=4)

# boolean operations
s = square(size=4);  c = circle(radius=3)
s | c    # union
s - c    # difference
s & c    # intersection
s ^ c    # symmetric difference
s + c    # concat (no merge)

# transforms
s.chamfer(size=0.8)              # vertex_filter=None, epsilon=1e-8
s.color(color=RED)
s.convexify()                    # part_filter=None
s.fill()                         # part_filter=None
s.fillet(size=0.8)               # vertex_filter=None, segment_count=32, epsilon=1e-8
s.grow(distance=0.5)             # part_filter=None
s.linear_cut(axis=x())          # pivot=0
s.linear_pattern(counts=4, steps=x(3))        # counts=[nx, ny], steps=[sx, sy]
s.linear_slice(thickness=2, direction=x())  # pivot=0, part_filter=None
s.mirror(normal=[1, 0])          # pivot=0
s.pull(distance=1.0)             # pivot=0, vertex_filter=None
s.push(distance=1.0)             # pivot=0, vertex_filter=None
s.radial_pattern(count=6)        # angle=360, pivot=0
s.radial_slice(start=0, end=180) # pivot=0, part_filter=None
s.resize(size=[6, None])         # auto=False, pivot=None, vertex_filter=None
s.rotate(angle=30)               # pivot=0, vertex_filter=None
s.scale(scale=[2, 0.5])          # pivot=0, vertex_filter=None
s.shrink(distance=0.5)           # part_filter=None
s.translate(translation=[2, 1])  # vertex_filter=None

# features
s.bounds                         # [min_x, min_y, max_x, max_y]
s.bounding_box                   # → Shape (rectangle)
s.centroid                       # [cx, cy] — geometric centroid
s.is_empty                       # bool

# topology — coordinates & attributes
s.are_vertices_convex            # (n_vertices,)   — convexity mask
s.directed_edge_directions       # (2*n_edges, 2)
s.edge_lengths                   # (n_edges,)
s.edge_midpoints                 # (n_edges,  2)
s.edge_normals                   # (n_edges,  2)
s.ring_types                     # (n_rings,)  — "exterior"|"interior"
s.vertex_angles                  # (n_vertices,)   — interior angles (°)
s.vertex_coordinates             # (n_vertices, 2)
s.vertex_normals                 # (n_vertices, 2) — outward unit normals

# topology — bridges (*_to_*)
s.directed_edge_to_edge             # directed_edge → edge
s.directed_edge_to_vertex           # directed_edge → [start, end]
s.edge_to_vertex                    # edge          → [start, end]
s.ring_to_part                      # ring          → part
s.vertex_to_incoming_directed_edge  # vertex        → directed_edge
s.vertex_to_outgoing_directed_edge  # vertex        → directed_edge
s.vertex_to_neighbor_vertex       # vertex        → [prev, next]
s.vertex_to_part                    # vertex        → part
s.vertex_to_ring                    # vertex        → ring

# extrusions → Solid
s.linear_extrude(height=3)
s.radial_extrude(axis=y(), pivot=x(5))  # start=0, end=360, segment_count=64
s.path_extrude(path)                    # fillet_segments=None, min_fillet_radius=None, intermediate_sections=None, strategy=None

# sweep strategies (for path_extrude strategy= parameter)
scale_sweep(end=3)                      # start=1.0
rotate_sweep(angle=360)                 # start_angle=0.0
resize_sweep(end_size=[2, 4])           # start_size=None
reverse_sweep(strategy=scale_sweep(3))

# export
s.to_dxf_file("output.dxf")
s.to_html_file("output.html")
s.to_screen()
s.to_svg_file("output.svg")

3D — Solid

from scadpy import *

# primitives
cone(radius=2, height=4)         # section_count=32
cuboid(size=[4, 3, 2])
cylinder(radius=2, height=4)     # section_count=32
polyhedron(vertices=vertices, faces=faces)
sphere(radius=3)                 # subdivision_count=4
Solid.from_stl("model.stl")

# boolean operations
a = cuboid(size=[4, 3, 2]);  b = sphere(radius=2)
a | b    # union
a - b    # difference
a & b    # intersection
a ^ b    # symmetric difference
a + b    # concat (no merge)

# transforms
a.color(color=RED)
a.convexify()                    # part_filter=None
a.linear_pattern(counts=4, steps=x(3))        # counts=[nx, ny, nz], steps=[sx, sy, sz]
a.mirror(normal=[1, 0, 0])       # pivot=0
a.pull(distance=1.0)             # pivot=0, vertex_filter=None
a.push(distance=1.0)             # pivot=0, vertex_filter=None
a.radial_pattern(count=6, axis=z())            # angle=360, pivot=0
a.resize(size=[6, None, None])   # auto=False, pivot=None, vertex_filter=None
a.rotate(angle=30, axis=z())    # pivot=0, vertex_filter=None
a.scale(scale=[2, 1, 0.5])       # pivot=0, vertex_filter=None
a.translate(translation=[1, 0, 0])  # vertex_filter=None

# features
a.bounds                         # [min_x, min_y, min_z, max_x, max_y, max_z]
a.bounding_box                   # → Solid (cuboid)
a.centroid                       # [cx, cy, cz] — geometric centroid
a.is_empty                       # bool

# topology — coordinates & bridges (*_to_*)
a.triangle_to_vertex    # triangle → [v0, v1, v2]
a.vertex_coordinates    # (n_vertices,  3)
a.vertex_to_part        # vertex   → part

# export
a.to_html_file("output.html")
a.to_screen()
a.to_stl_file("output.stl")

Roadmap

  • Improve documentation
  • Richer topology for Shape and Solid
  • Richer transformations for Shape and Solid
  • Chamfer and fillet on Solid
  • New assembly types: PointCloud2d, Wire2d, PointCloud3d, Wire3d
  • Better error messages
  • More import/export formats

Development

# Create and activate venv
python3 -m venv .venv
source .venv/bin/activate

# Install with dev dependencies
pip install -e .[dev]

# Run doctests & generate documentation & AI skill file
cd docs && make doctest && make html && make skills

License

See LICENSE.md.

Project details


Download files

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

Source Distribution

scadpy-0.4.24.tar.gz (485.1 kB view details)

Uploaded Source

Built Distribution

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

scadpy-0.4.24-py3-none-any.whl (201.6 kB view details)

Uploaded Python 3

File details

Details for the file scadpy-0.4.24.tar.gz.

File metadata

  • Download URL: scadpy-0.4.24.tar.gz
  • Upload date:
  • Size: 485.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for scadpy-0.4.24.tar.gz
Algorithm Hash digest
SHA256 9c54a829661450bb99119be5edc8125ab92fa58046e5ce80254064e7eaef1222
MD5 9ab9ed60a916a2d54f810e0a94efe91d
BLAKE2b-256 aefb8333364bc2c2cec88514988c1e4ecf263140bcb4d596a0db4cd08085bbbc

See more details on using hashes here.

Provenance

The following attestation bundles were made for scadpy-0.4.24.tar.gz:

Publisher: release.yml on m-fabregue/scadpy

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

File details

Details for the file scadpy-0.4.24-py3-none-any.whl.

File metadata

  • Download URL: scadpy-0.4.24-py3-none-any.whl
  • Upload date:
  • Size: 201.6 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for scadpy-0.4.24-py3-none-any.whl
Algorithm Hash digest
SHA256 8eb6c994e0321836eb0ae76569e7b19da0939ccb16643383add440dcd315a6c2
MD5 eb00f96d661f41deeb9dcd1f4f8b1d81
BLAKE2b-256 e420238093e68f4785ef6f27007a7ee9d69320d8b8acd557ef39f917189cfcc9

See more details on using hashes here.

Provenance

The following attestation bundles were made for scadpy-0.4.24-py3-none-any.whl:

Publisher: release.yml on m-fabregue/scadpy

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

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page