Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/ekampf/pymaybe

A Python implementation of the Maybe pattern
https://github.com/ekampf/pymaybe

Last synced: 2 days ago
JSON representation

A Python implementation of the Maybe pattern

Awesome Lists containing this project

README

        

===============================
PyMaybe
===============================

.. image:: https://travis-ci.org/ekampf/pymaybe.svg?branch=master
:target: https://travis-ci.org/ekampf/pymaybe

.. image:: https://coveralls.io/repos/ekampf/pymaybe/badge.svg?branch=master&service=github
:target: https://coveralls.io/github/ekampf/pymaybe?branch=master

.. image:: https://img.shields.io/pypi/v/pymaybe.svg
:target: https://pypi.python.org/pypi/pymaybe

.. image:: https://img.shields.io/pypi/dm/pymaybe.svg
:target: https://pypi.python.org/pypi/pymaybe

A Python implementation of the Maybe pattern.

Installation
------------

.. code::

pip install pymaybe

Getting Started
---------------

.. code::

from pymaybe import maybe
first_name = maybe(deep_hash)['account']['user_profile']['first_name'].or_else("")

Documentation
-------------
Maybe monad is a programming pattern that allows to treat None values that same way as non-none values.
This is done by wrapping the value, which may or may not be None to, a wrapper class.

The implementation includes two classes: *Maybe* and *Something*.
*Something* represents a value while *Nothing* represents a None value.
There's also a method *maybe* which wraps a regular value and and returns *Something* or *Nothing* instance.

.. code::

>>> maybe("I'm a value")
"I'm a value"

>>> maybe(None);
None

Both *Something* and *Nothing* implement 4 methods allowing you to test their real value: *is_some*, *is_none*, *get* and *or_else*

.. code::

>>> maybe("I'm a value").is_some()
True

>>> maybe("I'm a value").is_none()
False

>>> maybe(None).is_some()
False

>>> maybe(None).is_none()
True

>>> maybe("I'm a value").get()
"I'm a value"

>>> maybe("I'm a value").or_else(lambda: "No value")
"I'm a value"

>>> maybe(None).get()
Traceback (most recent call last):
...
Exception: No such element

>>> maybe(None).or_else(lambda: "value")
'value'

>>> maybe(None).or_else("value")
'value'

In addition, *Something* and *Nothing* implement the Python magic methods allowing you to treat them as dictionaries:

.. code::

>>> nested_dict = maybe(nested_dict)
>>> nested_dict['store']['name']
'MyStore'

>>> nested_dict['store']['address']
None

>>> nested_dict['store']['address']['street'].or_else('No Address Specified')
'No Address Specified'

All other method calls on *Something* are forwarded to its real *value*:

.. code::

>>> maybe('VALUE').lower()
'value'

>>> maybe(None).invalid().method().or_else('unknwon')
'unknwon'

Examples & Use Cases
--------------------

The Maybe pattern helps you avoid nasty try..except blocks.
Consider the following code:

.. code::

try:
url = rss.load_feeds()[0].url.domain
except (TypeError, IndexError, KeyError, AttributeError):
url = "planetpython.org"

With Maybe you could simply do:

.. code::

url = maybe(rss).load_feeds()[0]['url'].domain.or_else("planetpython.org")

Getting the current logged in user's name.
Without maybe:

.. code::

def get_user_zipcode():
address = getattr(request.user, 'address', None)
if address:
return getattr(address, 'zipcode', '')

return ''

With maybe:

.. code::

def get_user_zipcode():
return maybe(request.user).address.zipcode.or_else('')

Further Reading
---------------

* `Option (Scala) `_
* `Maybe (Java) `_
* `Maybe pattern (Python recipe) `_
* `Data.Maybe (Haskell) `_
* `Maybe (Ruby) `_

Copyright and License
---------------------
Copyright 2015 - `Eran Kampf `_

* Free software: BSD license
* Documentation: https://pymaybe.readthedocs.org.
* Code is hosted on `GitHub `_