Skip to content

The fastest and most intuitive library to manipulate STL files (stereolithography) for C++ and Python, header-only.

License

Notifications You must be signed in to change notification settings

Innoptech/OpenSTL

Repository files navigation

OpenSTL

The fastest and most intuitive library to manipulate STL files (stereolithography) for C++ and Python, header-only.

Commitizen friendly Conventional Commits PyPI license
pypi build Python

🌟 ✊ Please consider starring and sponsoring the GitHub repo to show your support! ✊ 🌟
GitHub Sponsor

Performances benchmark

Discover the staggering performance of OpenSTL in comparison to numpy-stl, meshio and stl-reader, thanks to its powerful C++ backend. See benchmark.py. Benchmark performed on an Intel i5-9600KF CPU @ 3.70GHz.

Performance gains over numpy-stl, meshio and stl-reader
Write:  1.3 to 4+ X faster
Read:   1 to 2.3+ X faster
Rotate: 1 to 12+  X faster

Note: meshio has no straightfoward way of rotating vertices, so it was not benchmarked. Benchmark Results

Python Usage

Install

pip install openstl or pip install -U git+https://github.com/Innoptech/OpenSTL@main

Read and write from a STL file

import openstl
import numpy as np

# Define an array of triangles
# Following the STL standard, each triangle is defined with : normal, v0, v1, v2
quad = np.array([
    # normal,          vertices 0,      vertices 1,      vertices 2
    [[0.0, 0.0, 1.0], [0.0, 0.0, 0.0], [1.0, 0.0, 0.0], [1.0, 1.0, 0.0]], # Triangle 1
    [[0.0, 0.0, 1.0], [0.0, 0.0, 0.0], [0.0, 1.0, 0.0], [1.0, 1.0, 0.0]], # Triangle 2
])

# Serialize the triangles to a file
success = openstl.write("quad.stl", quad, openstl.format.binary) # Or openstl.format.ascii (slower but human readable)

if not success:
    raise Exception("Error: Failed to write to the specified file.")

# Deserialize triangles from a file
deserialized_quad = openstl.read("quad.stl")

# Print the deserialized triangles
print("Deserialized Triangles:", deserialized_quad)

Rotate, translate and scale a mesh

import openstl
import numpy as np

quad = openstl.read("quad.stl")

# Rotating
rotation_matrix = np.array([
    [0,-1, 0],
    [1, 0, 0],
    [0, 0, 1]
])
rotated_quad = np.matmul(rotation_matrix, quad.reshape(-1,3).T).T.reshape(-1,4,3)

# Translating
translation_vector = np.array([1,1,1])
quad[:,1:4,:] += translation_vector # Avoid translating normals

# Scaling
scale = 1000.0
quad[:,1:4,:] *= scale # Avoid scaling normals

Convert Triangles ➡️ Vertices and Faces

import openstl

# Define an array of triangles
triangles = [
    # normal,          vertices 0,      vertices 1,      vertices 2
    [[0.0, 0.0, 1.0], [0.0, 0.0, 0.0], [1.0, 0.0, 0.0], [1.0, 1.0, 0.0]], # Triangle 1
    [[0.0, 0.0, 1.0], [0.0, 0.0, 0.0], [0.0, 1.0, 0.0], [1.0, 1.0, 0.0]], # Triangle 2
]

# Convert triangles to vertices and faces
vertices, faces = openstl.convert.verticesandfaces(triangles)

Convert Vertices and Faces ➡️ Triangles

import openstl

# Define vertices and faces
vertices = [
    [0.0, 0.0, 0.0],
    [1.0, 1.0, 1.0],
    [2.0, 2.0, 2.0],
    [3.0, 3.0, 3.0],
]

faces = [
    [0, 1, 2],  # Face 1 
    [1, 3, 2]   # Face 2 
]

# Convert vertices and faces to triangles
triangles = openstl.convert.triangles(vertices, faces)

C++ Usage

Read STL from file

#include <openstl/core/stl.h>

std::ifstream file(filename, std::ios::binary);
if (!file.is_open()) {
    std::cerr << "Error: Unable to open file '" << filename << "'" << std::endl;
}

// Deserialize the triangles in either binary or ASCII format
std::vector<openstl::Triangle> triangles = openstl::deserializeStl(file);
file.close();

Write STL to a file

std::ofstream file(filename, std::ios::binary);
if (!file.is_open()) {
    std::cerr << "Error: Unable to open file '" << filename << "'" << std::endl;
}

std::vector<openstl::Triangle> originalTriangles{}; // User triangles
openstl::serialize(originalTriangles, file, openstl::StlFormat::Binary); // Or StlFormat::ASCII

if (file.fail()) {
    std::cerr << "Error: Failed to write to file " << filename << std::endl;
} else {
    std::cout << "File " << filename << " has been successfully written." << std::endl;
}
file.close();

Serialize STL to a stream

std::stringstream ss;

std::vector<openstl::Triangle> originalTriangles{}; // User triangles
openstl::serialize(originalTriangles, ss, openstl::StlFormat::Binary); // Or StlFormat::ASCII

Convert Triangles ➡️ Vertices and Faces

using namespace openstl

std::vector triangles = {
    //        normal,             vertices 0,         vertices 1,        vertices 2
    Triangle{{0.0f, 0.0f, 1.0f}, {1.0f, 1.0f, 1.0f}, {2.0f, 2.0f, 2.0f}, {3.0f, 3.0f, 3.0f}},
    Triangle{{0.0f, 0.0f, 1.0f}, {2.0f, 2.0f, 2.0f}, {3.0f, 3.0f, 3.0f}, {4.0f, 4.0f, 4.0f}}
};

const auto& [vertices, faces] = convertToVerticesAndFaces(triangles);

Convert Vertices and Faces ➡️ Triangles

using namespace openstl

std::vector vertices = {
    Vec3{0.0f, 0.0f, 0.0f}, Vec3{1.0f, 1.0f, 1.0f}, Vec3{2.0f, 2.0f, 2.0f}, Vec3{3.0f, 3.0f, 3.0f}
};
std::vector<Face> faces = {
    {0, 1, 2}, {3, 1, 2}
};

const auto& triangles = convertToTriangles(vertices, faces);

Integrate to your codebase

Smart method

Include this repository with CMAKE Fetchcontent and link your executable/library to openstl::core library.
Choose weither you want to fetch a specific branch or tag using GIT_TAG. Use the main branch to keep updated with the latest improvements.

include(FetchContent)
FetchContent_Declare(
    openstl
    GIT_REPOSITORY https://github.com/Innoptech/OpenSTL.git
    GIT_TAG main
    GIT_SHALLOW TRUE
    GIT_PROGRESS TRUE
)
FetchContent_MakeAvailable(openstl)

Naïve method

Simply add stl.h to your codebase.

Test

git clone https://github.com/Innoptech/OpenSTL
mkdir OpenSTL/build && cd OpenSTL/build
cmake -DOPENSTL_BUILD_TESTS=ON .. && cmake --build .
ctest .

Requirements

C++11 or higher.

DISCLAIMER: STL File Format

The STL file format, while widely used for 3D modeling and printing, was designed to be simple and easy to parse. However, this simplicity comes with some significant limitations:

  • Lack of Built-in Validation Mechanisms: The STL format does not include built-in mechanisms such as checksums, hashes, or any form of file validation. This makes it challenging to detect certain types of file corruption, such as a truncated header or malformed data. As a result, errors in file transmission, storage, or manipulation might go undetected.

  • Vulnerability to Corruption: Due to the lack of validation features, STL files can be easily corrupted. For example, if the file is truncated or contains invalid data, these issues may not be detected until the file is parsed or processed, potentially leading to crashes or undefined behavior in applications that use the file.

  • Potential for Buffer Overflow Attacks: The lack of built-in validation and the absence of bounds checking in the STL format can make it susceptible to buffer overflow attacks. Care should be taken when handling STL files, especially those from untrusted sources, to ensure they are properly validated before being used.

These limitations are inherent to the STL format and should be considered when working with or implementing software that processes STL files. Developers are encouraged to implement additional validation and error-handling mechanisms in their applications to mitigate these risks.