https://github.com/huy-dna/pryvacy

Access control decorators for python classes
https://github.com/huy-dna/pryvacy

decorator emulation encapsulation oop poetry-python python

Last synced: 22 days ago
JSON representation

Access control decorators for python classes

• Host: GitHub
• URL: https://github.com/huy-dna/pryvacy
• Owner: Huy-DNA
• License: apache-2.0
• Created: 2024-05-29T05:14:32.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2024-06-02T16:57:03.000Z (5 months ago)
• Last Synced: 2024-06-02T18:45:52.248Z (5 months ago)
• Topics: decorator, emulation, encapsulation, oop, poetry-python, python
• Language: Python
• Homepage:
• Size: 64.5 KB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGELOG.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# pryvacy - a set of access control decorators for python

[![PyPI Version](https://img.shields.io/pypi/v/pryvacy.svg)](https://pypi.python.org/pypi/pryvacy)

## Philosophy

* Incur as least overhead as possible when using the access control decorators

* Only classes that use the decorators need to be changed, other related classes do not need to -> Can be opted-in easily

## Installation

```bash

pip install pryvacy

```

## Usage

The package provides 3 access control decorators: `@private`, `@public`, `@protected` that can be used on methods and nested classes (currently only `@private` can be used on nested classes)

```python

from pryvacy import pryvacy, private, public, protected

@pryvacy

class Foo():

  @public

  def public_method(self):

    pass

  @protected

  def protected_method(self):

    pass

  @private

  def private_method(self):

    pass

```

Access control rules:

* Methods inside `Foo` are able to access `public_method`, `protected_method`, `private_method`.

* Code outside `Foo` and not inside any `Foo`'s subclass methods can only access `public_method`.

* Methods inside `Foo`'s subclasses (either decorated with `@pryvacy` or not) can access `public_method` and `protected_method`.

* Nested classes methods can access `public_method`, `protected_method`, `private_method`.

## Pitfalls & Bugs

Disclaimer: The package has not been tested thoroughly! Use with caution! Any contributions are appreciated~

1. Currently, class-level code cannot access `protected_method` and `private_method`.

   Example:

```python

    class Foo():

        ...

        class Bar(Foo):

            Foo().public_method() # OK!

            Foo().protected_method() # Exception!

            Foo().private_method() # Exception!

```

2. The `test_back_and_forth` test is failing. This scenario is unlikely to happen but I will try to come up with a performant solution.

## Current limitations

* `@private` and `@protected` are not supported on nested classes yet.

* No way to enforce access policy on class and instance attributes.

## Roadmap

* Benchmark decorated classes

* Test comprehensively the decorators interaction with the whole ecosystem

* Implement @private and @protected on nested classes