Skip to main content

Generates Python Extension modules from commented Cython PXD files

Project description

autowrap build autowrap | Conda Version

Generates Python Extension modules from Cython PXD files.

Introduction

One important application of Cython is to wrap C++ classes for using them in Python. As Cythons syntax is quite similar to the syntax of Python writing a wrapper can be learned easily. Further Cython prevents you from many typical errors which might get in your way if you write such a wrapper in C++.

Requirements

  • Python ≥ 3.9
  • Cython ≥ 3.0

This wrapping process typically consist of four steps:

  1. Rewrite parts of the header files of your C++ library in so called .pxd files. These give Cython information for calling the library and for error checking the code written in the following step.

  2. Write Cython code which wraps the C++ library. This code resides in one or more .pyx files.

  3. Translate these .pyx files to C++ code with Cython.

  4. Use distutils to compile and link the C++ code to the final Python extension module.

Depending on the size of your library step 2 can be tedious and the code will consist of many similar code blocks with only minor differences.

This is where autowrap comes into play: autowrap replaces step 2 by analyzing the .pxd files with Cythons own parser and generating correct code for step 3. In order to steer and configure this process the .pxd files can be annotated using special formatted comments.

The main work which remains is writing the .pxd files. This is comparable to the declarations you have to provide if you use SIP or SWIG.

Documentation

We assume that you installed autowrap already, so running

$ autowrap --help

does not fail.

Please see docs/README.md for further documentation.

Features

  • Wrapping of template classes with their public methods and attributes,enums, free functions and static methods.

  • Included converters from Python data types to (many) STL containers and back. As this is version 0.2, not all STL containers are supported. We plan full support of nested STL containers.

  • Manually written Cython code can be incorporated for wrapping code which autowrap can not handle (yet), and for enriching the API of the wrapped library. As this is done by writing Cython instead of C/C++ code, we get all benefits which Cython shows compared to C/C++.

    Writing a code generator for handling all thinkable APIs is hard, and results in a difficult and hard to understand code base. We prefer a maintainable code generator which handles 95% of all use cases, where the remaining 5% are still wrapped manually.

  • For achieving a pythonic API, converters for library specific data types can be implemented easily. These converters are written in Python and Cython, not in C/C++ code using the C-API of CPython.

  • autowrap relies on Cython, so we get automatic conversion of C++ exceptions to Python exceptions and wrapper code with correct reference counting. Using distutils we do not have to care to much about the build process on the targeted platform.

  • Support for generating some special methods, as __getitem__, __copy__ and numerical comparison operators.

Credits

Many thanks go to:

  • Hannes Roest, ETH Zürich, for contributing new ideas, patches, fruitful discussions and writing the first draft of this README.

  • Lars Gustav Malmström, ETH Zürich, for getting the ball rolling.

  • The developers of Cython for providing such a powerful and high quality tool.

  • Thanks to https://github.com/hendrik-cliqz for implementing the "no-gil" annotation.

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

autowrap-0.27.0.tar.gz (111.3 kB view details)

Uploaded Source

Built Distribution

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

autowrap-0.27.0-py3-none-any.whl (79.4 kB view details)

Uploaded Python 3

File details

Details for the file autowrap-0.27.0.tar.gz.

File metadata

  • Download URL: autowrap-0.27.0.tar.gz
  • Upload date:
  • Size: 111.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for autowrap-0.27.0.tar.gz
Algorithm Hash digest
SHA256 7f746e006a92409dd2cba04e033dd415ffe759ae5620a93a12ac734068430831
MD5 3cf85a16ae70b2e5b0da98915314aa1c
BLAKE2b-256 4c093c50695d39b8245ce620d0840358e2e1ecac76797693b7f8e16fbe497a46

See more details on using hashes here.

Provenance

The following attestation bundles were made for autowrap-0.27.0.tar.gz:

Publisher: release-autowrap.yaml on OpenMS/autowrap

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

File details

Details for the file autowrap-0.27.0-py3-none-any.whl.

File metadata

  • Download URL: autowrap-0.27.0-py3-none-any.whl
  • Upload date:
  • Size: 79.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for autowrap-0.27.0-py3-none-any.whl
Algorithm Hash digest
SHA256 a4d3b0aef4da1ab7d075d8fc3e679b1ff18a25a0c09bdb551d5c770d867f3ac2
MD5 f475bc984106b7b31610a9010908b5dc
BLAKE2b-256 5e90e36ae678ffac2be7ef81811754ca1be1deede2f0d010055d148837e5f498

See more details on using hashes here.

Provenance

The following attestation bundles were made for autowrap-0.27.0-py3-none-any.whl:

Publisher: release-autowrap.yaml on OpenMS/autowrap

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