odin 1.0

Data-structure definition/validation/traversal, mapping and serialisation toolkit for Python

Odin provides a declarative framework for defining resources (classes) and their relationships, validation of the fields that make up the resources and mapping between objects (either a resource, or other python structures).

Odin also comes with built in serialisation tools for importing and exporting data from resources.

Highlights

• Class based declarative style

• Fields for building composite resources

• Field and Resource level validation

• Easy extension to support custom fields

• Python 2.7 ¹, Python 2.7+, Python 3.4+ and PyPy ¹ supported

• Integration with Django (see baldr)

• Support for documenting resources with Sphinx

• Minimal dependencies (base functionality only requires six)

¹ certain contrib items are not supported. Pint is not installable with PyPy.

Use cases

• Design, document and validate complex (and simple!) data structures

• Convert structures to and from different formats such as JSON, YAML, MsgPack or CSV

• Validate API inputs

• Define message formats for communications protocols, like an RPC

• Map API requests to ORM objects

Quick links

• Documentation

• Project home

• Issue tracker

Upcoming features

In development

• XML Codec (export completed)

• Complete documentation coverage

• Improvements for CSV Codec (writing, reading multi resource CSV’s)

• RESTful interface with support for Flask and Django

• Integration with other libraries (ie Django Models/Forms)

• Integration with SQLAlchemy

Requires

• six

Optional

• simplejson - Odin will use simplejson if it is available or fallback to the builtin json library

• msgpack-python - To enable use of the msgpack codec

• pyyaml - To enable use of the YAML codec

Contrib

• jinja2 >= 2.7 - For documentation generation

• pint - Support for physical quantities using the Pint library.

Development

• pytest - Testing

• pytest-cov - Coverage reporting

Example

With definition:

import odin

class Author(odin.Resource):
    name = odin.StringField()

class Publisher(odin.Resource):
    name = odin.StringField()

class Book(odin.Resource):
    title = odin.StringField()
    authors = odin.ArrayOf(Author)
    publisher = odin.DictAs(Publisher)
    genre = odin.StringField()
    num_pages = odin.IntegerField()

>>> b = Book(
        title="Consider Phlebas",
        genre="Space Opera",
        publisher=Publisher(name="Macmillan"),
        num_pages=471
    )
>>> b.authors.append(Author(name="Iain M. Banks"))
>>> from odin.codecs import json_codec
>>> json_codec.dumps(b, indent=4)
{
    "$": "Book",
    "authors": [
        {
            "$": "Author",
            "name": "Iain M. Banks"
        }
    ],
    "genre": "Space Opera",
    "num_pages": 471,
    "publisher": {
        "$": "Publisher",
        "name": "Macmillan"
    },
    "title": "Consider Phlebas"
}

Authors

Tim Savage

Special Mention

I would like to acknowledge the strong influence on the design of Odin Resources from the Django ORM and it’s notable contributor Malcolm Tredinnick. He was a valued colleague who’s untimely passing left a large void in our company and the wider community.