Skip to main content

HookMan is a python package that provides a plugin management system to applications, specially those who are written (in totally or partially) in C++.

Project description

Hookman

https://img.shields.io/pypi/v/python-hookman.svg https://img.shields.io/conda/vn/conda-forge/python-hookman.svg https://img.shields.io/pypi/pyversions/python-hookman.svg https://codecov.io/gh/ESSS/hookman/branch/master/graph/badge.svg https://github.com/ESSS/hookman/workflows/Hookman%20-%20CI/badge.svg Documentation Status

This documentation covers HookMan usage & API.

For information about HookMan, read the section above. For public changelog and how the project is maintained, please check the GitHub page

What is HookMan?

HookMan is a python package that provides a plugin management system to applications, specially those who are written (in totally or partially) in C++.

It enables external contributors to implement plugins which act as extensions written in C/C++ that interact with the application through well-defined hooks.

This system was largely inspired by pluggy, the plugin system which powers pytest, tox, and devpi, but with the intent to be called from a C++ application rather than from Python.

It was conceived to facilitate the application development, allowing hooks to be exposed in a clear way and allowing plugins to be developed without access to classes or data from the application.

With HookMan your application can have access to the hooks implemented on plugins as simple as the example below.

# Initializing a class
hm = HookMan(specs=acme_specs, plugin_dirs=['path1','path2'])

hook_caller = hm.get_hook_caller()

# Getting access to the hook implementation
friction_factor = hook_caller.friction_factor()
env_temperature = hook_caller.env_temperature()

# Checking if the hook was implemented
assert friction_factor is not None
assert env_temperature is None

# Executing the hook, wherever it is implemented either in plugin A or B.
ff_result = friction_factor(1, 2.5)
env_tmp_result = env_temperature(35.5, 45.5)

How does it work?

In order to use HookMan in your application, it is necessary to inform which Hooks are available to be implemented through a configuration object.

With this configuration defined, users can create plugins that implement available Hooks extending the behavior of your application.

All plugins informed to your application will be validated by HookMan (to check which hooks are implemented), and an object holding a reference to the Hooks will be passed to the application.

The HookMan project uses the library pybind11 to interact between Python and C/C++, allowing a straightforward usage for who is calling the function (either in Python or in C++).

Defining some terminologies:

  • Application ⇨ The program that offers the extensions.

  • Hook ⇨ An extension of the Application.

  • Plugin ⇨ The program that implements the Hooks.

  • User ⇨ The person who installed the application.

Read the docs to learn more!

Credits

Thanks for pluggy, which is a similar project (plugin system) and source for many ideas.

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

History

0.5.0 (2023-02-10)

  • Allow to add some extra text to created hmplugin file.

0.4.0 (2020-10-23)

  • When removing plugins they are first moved to a .trash dir and not directly deleted.

  • Allow HookManager to call hooks of a specific plugin.

0.3.0 (2019-12-16)

  • Rename the parameter dst_path to dest_path on install_plugin method.

  • install_plugin now returns the name of the plugin when the installation is successful.

  • Now the library path dir is added to PATH environment variable before load the library (Only on Windows).

  • Added an optional “extras” entry to plugin definition yaml:

    • “extras” is a dictionary for adding (key, value) customized options, accessible in PluginInfo.extras;

    • Plugin generation accepts a dict of default (key, value) pairs to be added to extras;

0.2.0 (2019-02-08)

  • Moved load hook function code to HookCaller.load_impls_from_library function implemented in C++. This enables using hook functionality in projects which don’t use Python as their entry point.

  • HookSpecs now accepts an extra_includes argument, which can be used to add custom #include directives to the generated HookCaller.hpp file.

  • HookCaller now contains a std::vector of functions bound to plugin implementations. This allows multiple plugins to implement the same hook; how the results of each call is to behave is responsibility of the caller.

    Because of this, the following classes/methods have been removed because they are no longer relevant:

    • ConflictBetweenPluginsError

    • ConflictStatus

    • HookMan.ensure_is_valid

    • HookMan.get_status

  • Generated files now sport a “do not modify” comment header.

  • Generation of the bindings code for HookCaller is skipped if specs.pyd_name is not defined.

  • Code generation is now available directly in the command-line through the commands:

    • python -m hookman generate-plugin-template

    • python -m hookman generate-project-files

    • python -m hookman generate-hook-specs-h

    • python -m hookman package-plugin

  • Explicitly declare extern "C" calling convention in the hook_specs.h file.

  • The INIT_HOOKS macro has been removed as it didn’t have any useful function.

0.1.7 (2018-08-23)

  • First Release on PyPI.

0.1.6 (2018-08-23)

  • Never released, deployment error.

0.1.5 (2018-08-23)

  • Never released, deployment error.

0.1.4 (2018-08-23)

  • Never released, deployment error.

0.1.3 (2018-08-23)

  • Never released, deployment error.

0.1.2 (2018-08-23)

  • Never released, deployment error.

0.1.1 (2018-08-23)

  • Never released, deployment error.

  • Dropping bumperversion and using setuptool_scm

0.1.0 (2018-08-23)

  • Never released, deployment error.

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

python-hookman-0.5.0.tar.gz (48.7 kB view details)

Uploaded Source

Built Distribution

python_hookman-0.5.0-py3-none-any.whl (22.0 kB view details)

Uploaded Python 3

File details

Details for the file python-hookman-0.5.0.tar.gz.

File metadata

  • Download URL: python-hookman-0.5.0.tar.gz
  • Upload date:
  • Size: 48.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.9.16

File hashes

Hashes for python-hookman-0.5.0.tar.gz
Algorithm Hash digest
SHA256 6f009e6b944af847425121aa147483652c79f26c81ac51da1db0f98811681562
MD5 75b6277caa1c3d6ba4f4feb9be7e7a35
BLAKE2b-256 c07d7b8ea79998e0a540c2f97ef62041b8e6d666a5718d2e06f99e2d6eafe8fe

See more details on using hashes here.

File details

Details for the file python_hookman-0.5.0-py3-none-any.whl.

File metadata

File hashes

Hashes for python_hookman-0.5.0-py3-none-any.whl
Algorithm Hash digest
SHA256 317d662d1b05078af0290e6b2e66be8aba2a096588f6c1c3993a119df82e6924
MD5 4e14a98d96f20d5b6bc43696a4bfbeff
BLAKE2b-256 69c279d18d6ad4e9a28dedc0de3b83b7f955ddef5ac280eb8b15d784ea74d709

See more details on using hashes here.

Supported by

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