Skip to content

Latest commit

 

History

History
83 lines (62 loc) · 2.43 KB

README.md

File metadata and controls

83 lines (62 loc) · 2.43 KB

CppTest with pybind11 Python bindings

This project demonstrates a C++ class CppTest with a derived Python class PyTest, using PyBind11 to create Python bindings for C++ code. The project provides access to an integer array and a vector of doubles, allowing Python users to manipulate the data efficiently.

Features

  • CppTest class: Contains an integer array n[3] and a std::vector<double> with 4 randomly generated values.
  • Functions to print the integer array and vector content from C++.
  • Python bindings via PyBind11 allow users to access C++ functions in Python, including:
    • Accessing a memory view of the integer array and double vector.
    • Viewing array/vector contents using pybind11::memoryview.

Installation

  1. Create a Virtual Environment:

    ./create_virtualenv

  2. Build the Project:

    ../../cbuild -t Release (or Debug)

Usage

After compiling the project, you can run the Python bindings.

  1. Activate the virtual environment (if not already activated):

    source ../../build/python_env/activate

  2. Run Python and test the bindings:

    python pybind11_add_p.py
./build/Release/main_static (or ./build/Debug/main_static)
  • The project uses PyBind11 to expose C++ classes and functions to Python.
  • The CppTest class contains an array n[3] and a vector random_doubles initialized with 4 random double values.
  • A derived PyTest class provides two main functions:
    • array_int(): Exposes the integer array as a pybind11::memoryview, which allows manipulation from Python.
    • vector_double(): Exposes the double vector as a pybind11::memoryview.

These bindings allow seamless access to C++ data structures in Python for high performance and efficient memory management.

Example output

When running the Python code to interact with the bindings, you should see output similar to:

int from cpp: 1
string from cpp: This is a string
default call: 2
call one function: 3
Test String
unpack n[0] 22
unpack n[0] 56
unpack n[0] 87
getter n[0] 22
getter n[0] v2 22
getter constexpr 4
unpack vector[0]: 0.796542984287846
unpack vector[1]: 0.18343478789336848
unpack vector[2]: 0.7796909976126613
unpack vector[3]: 0.5968501615800565

When running the c++ code, you should see output similar to:

Hello World!
7
Test String
22
Integer array n: 22 56 87 
Double vector: 0.796543 0.183435 0.779691 0.59685