bases: A Python library for Base-N encodings
============================================
.. image:: https://img.shields.io/badge/python-3.7+-green.svg
:target: https://docs.python.org/3.7/
:alt: Python versions
.. image:: https://img.shields.io/pypi/v/bases.svg
:target: https://pypi.python.org/pypi/bases/
:alt: PyPI version
.. image:: https://img.shields.io/pypi/status/bases.svg
:target: https://pypi.python.org/pypi/bases/
:alt: PyPI status
.. image:: http://www.mypy-lang.org/static/mypy_badge.svg
:target: https://github.com/python/mypy
:alt: Checked with Mypy
.. image:: https://readthedocs.org/projects/bases/badge/?version=latest
:target: https://bases.readthedocs.io/en/latest/?badge=latest
:alt: Documentation Status
.. image:: https://github.com/hashberg-io/bases/actions/workflows/python-pytest.yml/badge.svg
:target: https://github.com/hashberg-io/bases/actions/workflows/python-pytest.yml
:alt: Python package status
.. image:: https://img.shields.io/badge/readme%20style-standard-brightgreen.svg?style=flat-square
:target: https://github.com/RichardLitt/standard-readme
:alt: standard-readme compliant
Bases provides a customisable, parametric implementation of several common styles of Base-N encoding, covering all cases appearing in the `multibase specification <https://github.com/multiformats/multibase>`_ (except for proquints).
.. contents::
Install
-------
You can install the latest release from `PyPI <https://pypi.org/project/bases/>`_ as follows:
.. code-block:: console
$ pip install --upgrade bases
Usage
-----
We suggest you import bases as follows:
>>> import bases
Below are some basic usage examples, to get you started: for detailed documentation, see https://bases.readthedocs.io/
Base encoding objects
^^^^^^^^^^^^^^^^^^^^^
>>> from bases import base32
>>> base32
FixcharBaseEncoding(
StringAlphabet('ABCDEFGHIJKLMNOPQRSTUVWXYZ234567',
case_sensitive=False),
pad_char='=', padding='include')
Encoding
^^^^^^^^
>>> b = bytes([70, 98, 190, 187, 66, 224, 178])
>>> base32.encode(b)
'IZRL5O2C4CZA===='
Decoding
^^^^^^^^
>>> s = 'IZRL5O2C4CZA===='
>>> base32.decode(s)
b'Fb\xbe\xbbB\xe0\xb2'
>>> list(base32.decode(s))
[70, 98, 190, 187, 66, 224, 178]
Case/padding variations
^^^^^^^^^^^^^^^^^^^^^^^
>>> b = bytes([70, 98, 190, 187, 66, 224, 178])
>>> base32.encode(b)
'IZRL5O2C4CZA===='
>>> base32lower = base32.lower()
>>> base32lower
FixcharBaseEncoding(
StringAlphabet('abcdefghijklmnopqrstuvwxyz234567',
case_sensitive=False),
pad_char='=', padding='include')
>>> base32lower.encode(b)
'izrl5o2c4cza===='
>>> base32nopad = base32.nopad()
>>> base32nopad.encode(b)
'IZRL5O2C4CZA'
Case sensitivity variations
^^^^^^^^^^^^^^^^^^^^^^^^^^^
>>> s = 'IZRL5O2C4CZA===='
>>> base32lower.decode(s)
b'Fb\xbe\xbbB\xe0\xb2'
>>> base32lower_casesensitive = base32lower.with_case_sensitivity(True)
>>> base32lower_casesensitive.decode(s)
bases.encoding.errors.NonAlphabeticCharError: Invalid character 'I'
encountered for alphabet StringAlphabet('abcdefghijklmnopqrstuvwxyz234567').
Custom base encodings
^^^^^^^^^^^^^^^^^^^^^
>>> base4 = bases.make("0123", kind="zeropad-enc", block_nchars=4)
>>> base4
ZeropadBaseEncoding(StringAlphabet('0123'), block_nchars=4)
API
---
For the full API documentation, see https://bases.readthedocs.io/
Contributing
------------
Please see `<CONTRIBUTING.md>`_.
License
-------
`MIT © Hashberg Ltd. <LICENSE>`_