Quickstart

Requirements

djes requires:

• Django >= 1.8
• elasticsearch-dsl-py >= 0.0.4
• Python 2.7, 3.3 or 3.4

Installation

You can install djes from PyPi:

$ pip install djes

Next, just add djes to your INSTALLED_APPS:

INSTALLED_APPS += (
    'djes',
)

By default, djes uses localhost:9200 as the location of your Elasticsearch server, but this can be customized using the ES_CONNECTIONS settings attribute.

ES_CONNECTIONS = {
    "default": {
        "hosts": "192.168.1.143:9200"
    }
}

Making Your Models Indexable

In order to make a model Indexable, just extend from Indexable, instead of Django's model.Model, like so:

from djes.models import Indexable

class SimpleObject(Indexable):
    foo = models.IntegerField()

Syncing Your Mappings

Now that youve setup some models, save the mappings to Elasticsearch, using the mangement command sync_es:

python manage.py sync_es

Searching

In order to use Elasticsearch, instead of your database, just use the search_objects manager, like so:

>>> SimpleObject.objects.create(foo=666)
<SimpleObject: SimpleObject object>
>>> SimpleObject.search_objects.search().filter('term', foo=666).count()
1
>>> SimpleObject.search_objects.search().filter('term', foo=666)[0]
<SimpleObject_ElasticSearchResult: SimpleObject_ElasticSearchResult object>
>>> SimpleObject.search_objects.search().filter('term', foo=666)[0].foo
666

This manager returns "shallow" versions of your Django model. Specifically, it might leave off some fields (depending or your indexing), and the save() method will be unavilable. If you want to get full Django objects, you can chain the .full() method, like so:

>>> SimpleObject.search_objects.search().filter('term', foo=1).full()[0]
<SimpleObject: SimpleObject object>

Note that this will have a performance impact, as you are performing an Elasticsearch query, and then at least one database query.