djangoes

Integrates ElasticSearch and Django. Not an ElasticSearch based ORM.

Features

djangoes wraps the Python client for ElasticSearch with a connections handler and gives an interface to configure and use it for production and testing purpose.

• Thread-safe connection handler that mimics Django's DATABASES handling
• Uses official ElasticSearch Python client.
• Connections and indices can be configured separately.
• Integration with Django's native tests and py.test

Please read the documentation, and do not hesitate to send feedback on GitHub.

Compatibility

The current version of djangoes works only with Python 3.4 and ElasticSearch server >= 1.3.

Django 1.6 and 1.7 are tested, and elasticsearch-py 1.3.0 is used.

There are no plans to support older versions of Python, Django, or ElasticSearch.

Install

Use pip to install the latest version of djangoes and its dependencies:

$ pip install django elasticsearch djangoes

If you are not using a virtualenv, you may need to use pip3 and not pip - it depends on your Linux distribution and configuration.

License

CC0, or Public Domain when it is possible and equivalent (it's kind of complicated for some countries). If you are not sure: CC0.

Contribute

djangoes is distributed under the CC0 license, so anyone can contribute: issues, PR, ideas, proposals, congrats, and tests.

If you want to submit a PR, please also think of the following:

• PEP8 compliant source code with docstrings,
• documentation,
• unit tests.

I'm working on making automated tests on the code that perform requests to ElasticSearch, so the current situation is not perfect yet.

Development environment

If you'd like to contribute, you will need to install ElasticSearch and create a virtualenv. In Ubuntu, you can install ElasticSearch using the provided repository.

Then you can use virtualenvwrapper to create the virtualenv and git to clone the GitHub repository:

$ mkproject --python=/usr/bin/python3.4 djangoes
$ cd djangoes
$ git clone git@github.com:exirel/djangoes .

Use pip to install dependencies:

$ pip install -r requirements.txt

Then you can run the tests:

$ make test

Or compile a coverage and pylint report:

$ make report

Or, if you are lazy like me:

$ make all

You can build the doc too:

$ make doc
$ xdg-open docs/_build/html/index.html