skip to navigation
skip to content

Not Logged In

elasticmodels 0.0.6

Elasticmodels helps you index and query your Django models using elasticsearch

Latest Version: 0.0.10

# Elasticmodels

Elasticmodels helps you index and query your Django models using elasticsearch.
It is designed to be an alternative to django-haystack when you need more control over
your index creation, and you are always going to use elasticsearch.

# Install

pip install elasticmodels

# Setup


In your Django settings file, define these variables:


"urls": ["http://localhost:9200/"],
"indexes": ["the_name_of_your_es_index"],
# "http_auth": "username:password",

# these are used when your index is created
"settings": {
"analysis": {
"analyzer": {
"snowball": {
"type": "snowball",
"stopwords": "_none_"

Add elasticmodels to INSTALLED_APPS:



## app/

In a Django app, create a file, like so:


from elasticmodels import Indexable
from django.template.loader import render_to_string
from .models import File, FileTag

class FileIndex(Indexable):
# specify the model class this index is for
model = File

def mapping(self):
Return the elasticsearch mapping for this model type
return {
"properties": {
"pk": {"type": "integer", "index": "not_analyzed"},
"content": {"type": "string", "analyzer": "snowball"},
"tags": {"type": "string", "analyzer": "keyword"},
"org_id": {"type": "integer", "index": "not_analyzed"},
"type": {"type": "integer", "analyzer": "keyword"},
"uploaded_by_id": {"type": "integer", "analyzer": "keyword"},

def prepare(self, obj):
Return obj transformed into a dict that corresponds to the mapping
you defined. This is what will be indexed by elasticsearch.
return {
"content": render_to_string("files/search.txt", {"object": obj}),
"tags": [ for ft in FileTag.objects.filter(file=obj).select_related("tag")],
"org_id": obj.org_id,
"type": obj.type,
"uploaded_by_id": obj.uploaded_by_id,

# Usage

## Deleting and recreating your index

./ rebuild_index

**This will delete the entire elasticsearch index** and recreate it. All your
model objects will be re-indexed.

## Adding an individual object to the index


from elasticmodels import make_searchable

f = File(name="Foo", type=1)



## Querying

Your subclass of elasticmodels.Indexable has a class attribute called `objects`
which returns an elasticutils `S` instance. You can then use whatever methods are
available in elasticutils on the S instance.



from elasticutils import F
from .search_indexes import FileIndex

results = FileIndex.objects.filter(F(type=1) | F(type=2)).query(content__match="foo")
for result in results:
print, result.content
File Type Py Version Uploaded on Size
elasticmodels-0.0.6.tar.gz (md5) Source 2014-07-28 5KB
  • Downloads (All Versions):
  • 12 downloads in the last day
  • 187 downloads in the last week
  • 869 downloads in the last month