Django Opensearch DSL is a package that allows the indexing of Django models in opensearch. It is built as a thin
wrapper around opensearch-py
so you can use all the features developed by the opensearch-py
team.
You can view the full documentation at https://django-opensearch-dsl.readthedocs.io.
- Based on
opensearch-py
so you can make queries with theSearch
object. - Management commands for creating, deleting, and populating indices and documents.
- Opensearch auto mapping from Django models fields.
- Complex field type support (
ObjectField
,NestedField
). - Index fast using
parallel
indexing.
django-opensearch-dsl
only support the supported version of each dependency (mainstream & lts).
Python
(supported versions)Django
(supported versions)opensearch-py
(compatibility matrix)
The easiest way to install django-opensearch-dsl
is through pip
:
pip install django-opensearch-dsl
Then add django_opensearch_dsl
to your INSTALLED_APPS
settings.
You must then define OPENSEARCH_DSL
in your Django settings.
For example:
OPENSEARCH_DSL = {
'default': {
'hosts': 'localhost:9200'
},
'secure': {
'hosts': [{"scheme": "https", "host": "192.30.255.112", "port": 9201}],
'http_auth': ("admin", "password"),
'timeout': 120,
},
}
OPENSEARCH_DSL
is then passed
to opensearchpy.connection.connections.configure
.
To index instances of the following model :
# models.py
class Car(models.Model):
name = models.CharField()
color = models.CharField()
description = models.TextField()
type = models.IntegerField(choices=[
(1, "Sedan"),
(2, "Truck"),
(4, "SUV"),
])
First create a subclass of django_opensearch_dsl.Document
containing the subclasses Index
(which define the index' settings) and Django
(which contains settings related to your django Model
). Finally,
register the class using registry.register_document()
decorator.
It is required to define Document
classes inside a file named documents.py
in your apps' directory.
# documents.py
from django_opensearch_dsl import Document
from django_opensearch_dsl.registries import registry
from .models import Car
@registry.register_document
class CarDocument(Document):
class Index:
name = 'cars' # Name of the Opensearch index
settings = { # See Opensearch Indices API reference for available settings
'number_of_shards': 1,
'number_of_replicas': 0
}
# Configure how the index should be refreshed after an update.
# See Opensearch documentation for supported options.
# This per-Document setting overrides settings.OPENSEARCH_DSL_AUTO_REFRESH.
auto_refresh = False
class Django:
model = Car # The model associated with this Document
fields = [ # The fields of the model you want to be indexed in Opensearch
'name',
'color',
'description',
'type',
]
# Paginate the django queryset used to populate the index with the specified size
# This per-Document setting overrides settings.OPENSEARCH_DSL_QUERYSET_PAGINATION.
queryset_pagination = 5000