traitschema 1.2.0

Serializable schema using traits

traitschema
===========

.. image:: https://travis-ci.org/mivade/traitschema.svg?branch=master
:target: https://travis-ci.org/mivade/traitschema

.. image:: https://readthedocs.org/projects/traitschema/badge/?version=latest
:target: http://traitschema.readthedocs.io/en/latest/?badge=latest
:alt: Documentation Status

.. image:: https://codecov.io/gh/mivade/traitschema/branch/master/graph/badge.svg
:target: https://codecov.io/gh/mivade/traitschema

.. image:: https://img.shields.io/github/release/mivade/traitschema.svg
:target: https://github.com/mivade/traitschema

Create serializable, type-checked schema using traits_ and Numpy. A typical use
case involves saving several Numpy arrays of varying shape and type.

.. _traits: http://docs.enthought.com/traits/


Defining schema
---------------

.. note::

The following assumes a basic familiarity with the ``traits`` package. See
its `documentation <http://docs.enthought.com/traits/>`_ for details.

In order to be able to properly serialize data, non-scalar traits should be
declared as a ``traits.api.Array`` type. Example:

.. code-block:: python

import numpy as np
from traits.api import Array, String
from traitschema import Schema

class NamedMatrix(Schema):
name = String()
data = Array(dtype=np.float64)

matrix = NamedMatrix(name="name", data=np.random.random((8, 8)))

For other demos, see the ``demos`` directory.


Saving and loading
------------------

Data can be stored in the following formats:

* HDF5 via ``h5py``
* JSON via the standard library ``json`` module
* Numpy ``npz`` format

Multiple schema can be saved at once to a zip file via
:func:`traitschema.bundle_schema` and loaded with
:func:`traitschema.load_bundle`.